The Official SaaStr Podcast: SaaS | Founders | Investors

Ahead of SaaStr AI London on Dec 1-2 (See you there!) we’re taking a look back at some of our favorite sessions from our European events. It was so great when Olivier Pomel, founder CEO of Datadog, joined us as they crossed $100,000,000 ARR in a candid conversation it would be harder to do today post-IPO.

The First $100,000,000 ARR at Datadog: How Olivier Pomel Built a Customer-Centric Monitoring Giant

From zero lines of code to 700 employees and doubling revenue annually, Datadog CEO Olivier Pomel shares the counterintuitive strategies that built one of the most customer-obsessed companies in B2B SaaS

Olivier’s Top 5 Toughest Learnings

* You can’t be customer-focused if you’re sales-driven OR engineering-driven - Most companies fall into one trap or the other. Sales teams optimize for closing the next deal (short-term), while engineering teams build for the long-term without bridging back to customers. Customer-centricity requires daily vigilance against both.

* Closed alphas with “perfect customers” give terrible signal - Handpicking the best companies and best people for early access actually makes it harder to learn. Customers need to self-select when the timing is right for them. Open betas revealed infinitely more than curated alphas ever did.

* Month-to-month contracts are better than annual deals for learning - Every instinct (and investor) tells you to sell annual contracts. But monthly contracts force bad news to surface immediately instead of a year later. A year of going in the wrong direction is devastating for a young company.

* There’s no MVP for enterprise infrastructure - The conventional wisdom about shipping minimal products doesn’t apply when selling to enterprises who need comprehensive solutions. You need depth across many features before you’re minimally useful. It’s a continuum, not a single viable moment.

* Pricing conversations reveal product truth better than any metric - Putting a dollar amount on features focuses customers’ minds like nothing else. When customers say “I won’t pay for that,” you get brutally honest feedback about value. This friction is healthy and teaches you where to go next.

When Olivier Pomel and his co-founder started Datadog in 2010, they didn’t write a single line of code for the first six months. For two engineers itching to build, this took “some restraint,” as Olivier puts it. But this decision to obsessively listen before building became the foundation of a company that would redefine infrastructure monitoring and grow to 700+ employees while doubling in size every single year.

At SaaStr Europa, Olivier pulled back the curtain on how Datadog became one of the most customer-centric companies in enterprise software—and why being truly customer-focused requires constantly fighting against your natural instincts.

The Problem: When Great Teams Hate Each Other

The genesis of Datadog came from a painful problem Olivier and his co-founder experienced firsthand. Despite working together across four different companies, knowing each other extremely well, and building their teams from scratch with a “strict no a*****e policy,” they ended up in a familiar nightmare scenario two years in.

“We ended up with developers that hated operations, operations that hated developers, fingerpointing—all of the things that you can imagine,” Olivier explained.

The question became simple: Why don’t we give all of those teams the same viewpoint? How do we get them aligned and understanding their infrastructure the same way?

This became Datadog’s founding mission—bringing DevOps together and bridging the gap between development and operations teams. What they didn’t fully realize at the time was that this wasn’t just a nice-to-have feature. It was actually one of the KEY reasons why companies would migrate from legacy IT to the cloud.

“We ended up right in the center of it,” Olivier said. “Today the company is about 700 people. We’ve been doubling the size of the company every single year. It turns out everybody is moving to the cloud and everybody needs to understand what’s happening to their systems and applications.”

The Counterintuitive Truth About Customer-Centricity

Here’s the part most founders get wrong: you can’t be customer-focused if you let your company become sales-driven OR engineering-driven.

“Everybody wants to be customer-focused,” Olivier noted. “But most companies end up being either sales-driven or engineering-driven. If you want to be customer-focused, you can’t be either of those.”

The Sales-Driven Trap: Sales teams are phenomenal at figuring out what’s going to get a deal done. But very often, getting the next deal done is NOT what you want to do for the long run for your customers. Short-term thinking dominates.

The Engineering-Driven Trap: Let your engineering teams run on their own, and you’ll end up with organizations where people focus way too much on their solutions and way too much on the long term. You’ll struggle to bridge that gap back to the customer.

The solution? “It’s a struggle of every day to make sure that we go back to the customer and start from there.”

This wasn’t a conscious choice at first—it was forced upon them. Starting in New York (not the Bay Area), without Google or Facebook pedigrees, without millions in funding, and without a suite of VCs telling them they were geniuses, Datadog had no choice but to obsessively focus on the problem.

“We basically had to focus on the problem. We spent the first couple of years listening to customers and trying to understand what the problem was.”

The First Two Years: Why They Didn’t Write Code for 6 Months

When you’re engineers and you start a company, every instinct tells you to build. Olivier and his co-founder resisted this for six months—and discovered something remarkable.

When you don’t have anything to sell, everybody is super happy to talk to you.

“You’ll get hours and hours of really fantastic people at fantastic companies and they’ll spend all that time explaining to you what their problems are, what’s working, what’s not working for them. They’ll be extremely candid.”

This changes the moment you have something to sell. “Then you’re tainted. You have too much of a vested interest and you’re trying to push something, so people won’t open up so easily.”

After six months of research, they spent another six months building their first alpha. When they deployed it to a small number of customers, they noticed the product was “way too open-ended, way too general.”

But here’s where they made a critical discovery that contradicts conventional wisdom about customer selection.

The Open Beta Revelation: Let Customers Self-Select

Datadog initially ran a closed alpha, handpicking the best companies and the best people at those companies. “It was actually really, really hard to get a lot of signal from these customers,” Olivier admitted.

Then they opened it up to a wide beta—and everything changed.

“It’s a lot easier for users to self-select and start using your product than for you to understand for whom you’re going to be the right thing at the right time.”

You can’t predict when customers will have free bandwidth, when all the stars will align for deployment, or when your solution will be exactly what they need. Let them tell you by trying it.

This open approach became core to Datadog’s growth strategy and fed directly into their customer-centric philosophy.

Why Datadog Doesn’t Believe in MVPs

Ask most SaaS founders about MVPs and you’ll get textbook answers about shipping fast and iterating. Olivier has a different take.

“I don’t think there’s an MVP for what we do. I think it’s a myth,” he stated bluntly.

Here’s why: Datadog sells to enterprises. They sell a product that needs to monitor everything happening across entire infrastructures and applications. There’s a very large number of features customers rely on, and none of these features in themselves are revolutionary.

“You need to have them, otherwise you cannot be minimally useful. It’s more of a continuum—you keep adding those features, and at certain point you have enough of them that customers can start buying.”

The hard part? Figuring out which features matter most. “That’s why you have to go back to the customer and make sure you have enough of them so you can actually extract some signal from those conversations.”

For enterprise B2B infrastructure companies, the MVP concept often doesn’t apply. You need depth before customers will even consider you.

Scaling Customer-Centricity: The 10-100 Person Phase

When Datadog hit the market, they were about 10-15 people. The phase of reaching initial scale ran from there to just under 100 people. This is where most companies start to lose their customer focus—but it’s also where you can build the most powerful feedback mechanisms.

Strategy #1: Only Sell Month-to-Month Contracts

“We didn’t sell yearly deals for a very long time. We only sold month-to-month, meaning customers had the opportunity to churn all the time.”

This was structural, not accidental. If something is wrong with the product, if you don’t solve the right problem, if it’s not valuable enough—you know right away. Customers churn and you can have a conversation with them.

“If you start selling term deals, you’re going to have the bad news about a year later. By then you’ve wasted a year. You’ve gone in the wrong direction. You’ve made all these mistakes you could avoid otherwise.”

The lesson: Find structural ways to get bad news quickly. Don’t let long-term contracts mask product problems.

Strategy #2: Put Every Engineer on Support Rotation

Every single engineer at Datadog goes on a one-week support rotation throughout the year. Not just professional support engineers—every engineer in the company.

“Engineers are super happy when it starts, and they’re even happier when it ends,” Olivier joked. “But really what it gives is empathy for the customer.”

Engineers see what problems customers face day-to-day. They see the consequences of their design choices. Sometimes features that seemed like fantastic ideas turn out to be confusing from the customer’s perspective.

“It was a great idea, but we’re going to change it because it doesn’t work quite as well.”

Strategy #3: Bring Engineers to Conferences to Give Demos

Datadog does extensive event marketing because companies migrating to the cloud need to go to conferences to learn. They exhibit at these conferences, give lots of demos—and they bring engineers to do all of it.

“Everybody’s on a rotation and everybody’s going to get to one, two, three, five maybe of those conferences and spend every time a full day basically giving demos to potential customers or existing customers and answering questions.”

This builds empathy for customers and makes the whole engineering team more confident about the problems they’re solving and their relationship to end users.

But here’s the trick: Many engineers aren’t naturally comfortable seeking strangers’ attention at a booth for hours. The solution? Have other people on the team responsible for getting attention and introducing visitors to engineers. Then engineers can settle into a rhythm of demos and questions without the anxiety of having to attract people.

What Metrics Actually Matter (and Which Ones Don’t)

Here’s something counterintuitive from a company that processes 4-5 trillion customer metrics per day: Datadog isn’t incredibly metrics-driven for their own product.

“Most of the metrics are lagging indicators and they don’t convey all the nuance of the value that our customers are going to find with our product. So we don’t actually train ourselves to optimize to the metrics.”

What they DO watch:

* Volume of data and infrastructure customers are monitoring: A sign of the value they’re providing, because customers deploy them into more places

* Engagement: But this isn’t revolutionary

* Churn: Which is why they focused on month-to-month contracts early on

The philosophy: Don’t let metrics become a substitute for actual customer conversations and understanding.

Scaling to 700: When Everything Changes

At 700 people across 16 offices worldwide, some of the early strategies don’t work anymore. The executive team can’t meet with all customers all the time. The co-founders and CPO can’t do all the product work.

This is where Datadog made a critical hire: product managers.

But not just any product managers. They didn’t hire their first PM until they were over 100 people, and when they did, they defined the role very specifically.

“Product managers are not here to invent product. They’re here to spend the most time possible with the customers outside of the office.”

Their job:

* Understand what customers’ problems are

* Understand what they’re using, what they’re not using

* Understand what they’re paying for and why

* Get the best understanding possible of the problem and the value those problems represent

* Present the product and get feedback

“Their role is not to invent the product but to make sure it solves the right problem and has the right value.”

This is a fundamentally different view of product management than most companies have. PMs aren’t visionaries sitting in a room designing the future—they’re customer researchers and feedback conduits.

The Startup-Within-a-Startup Problem

When you start branching into new products or new markets, you’re suddenly back in startup mode—and it’s easy not to realize it.

“It’s easy to get used to winning, and then you expect everything’s just going to work this way,” Olivier explained.

Teams working on new products need to behave exactly as you would when you were a tiny company:

* Soliciting as much feedback as quickly as possible

* Going as wide as possible with betas

There’s tension here. You don’t want to taint the opinion of your existing product by releasing a new product that isn’t as polished. But if you want that new product to work, you need to go very wide very quickly.

Most importantly, you need to create a culture where people seek the bad news.

“It’s much better to go to a customer and ask them, ‘This new product—are you going to pay for it? You’ve tried it. Is it good enough? Are you going to buy it?’ It’s actually much better to hear from them, ‘No, I’m not going to buy it, and this is why,’ than to kick the can down the road and hope that maybe in 3 months, 6 months, 9 months they’ll change their mind.”

When you know what’s not working, it gives you a thread to pull. It tells you where you need to go.

The Counterintuitive Truth About Pricing

Here’s where customer-centricity gets tested: the pricing conversation. Customers want to spend as little as possible. You want them to pay as much as possible. How do you stay aligned?

First, agree on what kind of company you are. Olivier sees two types:

Type 1: Low-End Disruptors The major feature is that it’s cheaper. The dynamic with customers is they’ll push you to be cheaper and cheaper and to figure out what you can take out of your product so it can be as cheap as possible.

Type 2: High-Value Products You charge more for your product and you have to deliver more value for that. The cycle with your customer is to deliver more and more.

“You have to agree with your customer that they’re looking for high value, high impact—not as cheap as possible—because otherwise you’re never going to meet.”

Once you’ve agreed on that, you’re aligned in the mid and long term. Your customers want you to be successful. They want your company to be in business two years from now. They want you to ship new features. They really want you to be successful.

After that, there’s still friction—but that friction is how you learn about the value of your product and where you need to go next.

“Asking for money and putting a dollar amount on the product really focuses the mind for the customers. It helps you get really, really good feedback on where you need to go. In the long run, it’s a very healthy relationship to have.”

The key insight: See your customer as a partner. Frame it as “Let’s do this together. You want me to succeed and I want you to succeed, and the commitment on your side is to pay.”

How Olivier Maintains Culture at 700 People

Culture isn’t about writing down seven values on a wall (though Olivier says they’ll do that eventually as they continue to scale). Culture is really about who you hire, who you fire, and who you promote.

To maintain a customer-centric culture at scale, Datadog trains managers to care about the details. Olivier himself still sees every single customer complaint and support email that comes in.

“I don’t read most of them. I delete most of them right away. I just scan them very quickly. But what this does is it helps me pattern match. It gives me a sense of what’s actually happening, how people are actually reacting to the product, what they’re saying.”

He never acts on them directly. He never goes back to the customer or tries to solve the problem. But if he wants to affect change, he goes back through the management chain to make sure people get feedback and decide whether they need to change something.

The other critical element: Be super careful about how you talk about customers inside the company.

In sales, it can be tempting to talk about customers in manipulative ways as you try to manage them through a process. In engineering or product, you can think customers are making the wrong choices—”Why are they doing this stupid thing with our product?”

“The way we see it is if they’re doing a stupid thing with our product, it’s our fault. We let them do that. We made them think they should be doing that, or we didn’t explain it right.”

Always assume the customer is right. And even if you think they’re wrong, the fact that they’re thinking something wrong is itself a fact. It exists. You have to deal with it and you have to value it.

You can’t let yourself dismiss what you hear from customers.

The Acquisition Test: Optimizing for After the Deal

Datadog has made two acquisitions, both heavily tilted toward the tech platform, product, and team—not revenue streams to add to their sales engine.

“What we optimize for is what’s going to happen after we close the deal. Are people going to stick around? How long are they going to stick around? Are we going to be able to build on top of that team and turn that team of 20 we just acquired into a department of the company that’s going to have 200, 500 people?”

It’s all optimized on the fit and what happens after the acquisition—whether the companies share the same values.

The One Piece of Advice: Run Toward the Bad News

If there’s one principle that runs through everything Datadog does, it’s this:

“Really look for the bad news. Run toward the bad news.”

Whether you just shipped the product or you’re renewing a big customer, the wrong thing to do is show up in front of the customer hoping they won’t bring up the issue in the account. “Hey, maybe with a bit of luck we can get away with it.”

“Actually no—the first thing you do is you talk about it. You bring it up. Then you hear directly from the customer what they have to say about it. It actually goes a long way toward building the partnership, and it also helps you learn from the conversation with your customers.”

This applies at every stage:

* First six months: Don’t build—just listen to bad news about the market

* Alpha stage: Open it up to get bad news faster

* Launch phase: Only sell month-to-month so bad news comes immediately

* Scale phase: Put engineers in support so they hear bad news directly

* New products: Ask customers point-blank if they’ll pay for it

* Big renewals: Lead with the problems

The Bottom Line

Building a customer-centric company isn’t about being nice or having good intentions. It’s about building structural mechanisms that force bad news to surface quickly and loudly.

It’s about resisting the natural pull toward being sales-driven (short-term thinking) or engineering-driven (disconnected from reality).

It’s about spending six months not coding when you’re itching to build.

It’s about letting customers self-select rather than trying to control who uses your product.

It’s about putting engineers in uncomfortable situations where they have to face customers directly.

It’s about only selling month-to-month when your investors want annual deals.

It’s about asking customers if they’ll pay for something and being ready to hear “no.”

From zero to 700 people and doubling revenue every year, Datadog’s growth came from one counterintuitive principle: The best way to win is to constantly seek out why you might be losing.

“You have to the wrong thing to do is hope the customer won’t bring up the issue,” Olivier concluded. “The first thing you do is you talk about it. You bring it up. Then you hear directly from the customer what they have to say about it.”

That’s customer-centricity at scale.

Olivier’s Top 5 Mistakes (In His Own Words)

* Running a closed alpha with handpicked customers - “It was actually really, really hard to get a lot of signal from these customers.” Trying to control who got early access and picking the “perfect” companies backfired. Opening up the beta was when everything changed.

* Building the first alpha too open-ended and general - “We noticed that the product was not exactly what it needed to be. It was way too open-ended, way too general.” Even after six months of customer research, they still got the scope wrong initially.

* Not realizing how big the cloud opportunity would be - “We didn’t quite understand at the time that the cloud was going to be so big. We thought, hey, looks cool, why don’t we build it for companies moving to these new cloud environments.” They underestimated the magnitude of the shift they were riding.

* Not understanding that DevOps bridging was the killer feature, not just a feature - “We didn’t quite realize was that bringing DevOps together, bridging those two teams, was not just a feature of the new world. It was actually one of the key reasons why companies would be moving from legacy IT to the cloud.” They had the solution before they understood its full strategic value.

* Almost letting new products follow the old playbook instead of startup mode - “It’s easy to get used to winning and then you expect everything’s just going to work this way.” When launching new products at scale, they had to consciously fight against the temptation to skip the scrappy early-stage customer validation work that made them successful in the first place.

Want more insights on building customer-obsessed B2B + AI companies? Join us at the next SaaStr AI London Dec 1-2, where founders and CEOs share the unfiltered truth about scaling from $0 to $100M ARR and beyond.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

We’re back! Harry, Rory and Jason!

The venture capital playbook is broken. Not bent — broken. In the latest 20VC x SaaStr episode, Harry Stebbings, Jason Lemkin, and Rory O’Driscoll dissect why even Sequoia is making dramatic leadership changes, why seed investing at $50M pre-money might not work anymore, and what it actually takes to build venture returns in the age of AI.

This isn’t your typical venture conversation about “exciting trends.” This is three investors with $3+ billion in combined AUM telling you what’s actually working, what’s spectacularly failing, and why the old playbook from 2015-2022 is now a liability.

Key Takeaways

On Venture Capital Evolution:

* Sequoia’s leadership transition reflects broader industry truth: most VCs and executives from the last decade aren’t the right people for the next decade

* The pace of AI evolution means knowledge from 6 months ago is probably wrong; staying current requires dedicated time investment

* Partnerships are inherently dysfunctional when performance can’t tie to economics, creating inevitable internal tension

On AI Investment Strategy:

* Only three ways to win: (1) Attach to compute budgets, (2) Replace human headcount, or (3) Massively displace incumbents

* Using AI to “make your product better” no longer earns any kudos — that’s table stakes in 2024

* Co-pilots were the 2024 story that didn’t work; agents becoming actual team members is the 2026 opportunity

On Deal Dynamics:

* Getting into deals at 5-10M ARR requires top-decile metrics — there’s almost no middle class of fundable companies

* The quality and speed of competitive clones has increased dramatically, compressing the window for building moats

* Traditional seed defensibility is dead; founders must run faster and bet on scale creating the moat, not early product advantages

On Portfolio Construction:

* With increased variance in AI deals, diversification becomes more critical, not less

* Small fund sizes ($40-100M) with acceptance of dilution can generate superior returns (10x+) versus large funds maintaining ownership (5x)

* 80+ company meetings per week per partnership is one approach; building deep relationships with fewer founders is another

On Fundraising Process:

* The best fundraises don’t feel like processes — they’re cultivated over months with 3-4 investors ready before the data room opens

* Taking a term sheet immediately versus “running a process” depends on capital efficiency and relationship quality

* Founders often overlearn “run a process” advice without understanding the optimal approach is having everyone ready to commit before you formally raise

On Market Dynamics:

* Companies attached to AI compute infrastructure (like DataDog) are crushing it; those just using AI for product improvement (like Duolingo) are getting punished

* The $100M ARR milestone with 50 people (like Gamma) represents a new efficiency paradigm

* Capital-efficient outcomes (Billion to One at $5B, Navan at $4.5B) deliver superior investor returns despite smaller headline valuations

Sequoia’s Move: What It Really Means

When Sequoia replaced Roelof Botha as managing partner after just three years with Pat Grady and Alfred Lin in a split leadership structure, the venture world noticed. But the reaction from our panel was telling: this wasn’t about interpersonal drama or normal succession planning.

“Whenever you have a CEO change happening in venture, it’s because something isn’t working,” Rory stated bluntly. “This is dissatisfaction about how the firm is doing relative to the competition. They missed some rounds and some deals. They passed on some great companies.”

But rather than viewing this as a Sequoia-specific challenge, the panel sees it as symptomatic of a broader industry issue.

“More people should be stepping aside today,” Jason argues. “VCs, executives, founders. Most folks from the last decade or 15 years are not the right people for the next decade. I genuinely believe it across my ecosystem. I struggle to even recommend a lot of the CROs and executives I know for roles today.”

Rory adds context: “The pace of evolution is so fast. If you decide what I knew 6 months ago is still useful, you’re probably going to be wrong very quickly. That’s what I find the most stressful about right now.”

The meta-lesson? Even the best firms are struggling to keep pace with AI, which means everyone else should be asking themselves hard questions about whether they’re still the right people for this moment.

And Sequoia gets credit for one thing: ruthless decisiveness. “They did not do that fatal error of saying it’s someone’s turn so we’ll leave him in,” Rory notes. “They ruthlessly said, if we’re going to compete, we need these people, not those people.”

Michael Burry’s $1.1B Bet Against Nvidia: Why Shorting AI is Brutally Hard

Michael Burry made headlines (again) by shorting Nvidia and Palantir to the tune of $1.1 billion. Unlike most pundits who just discussed whether he’d be right, Jason actually did the math on what it takes to make money on these bets.

“It brings home how hard a business it is to bet against AI capex,” Jason concludes. “You not just got to be right, but you got to be right on timing.”

Rory’s take: “It’s easy to be roughly right. It’s very hard to imagine that the AI capex boom doesn’t have a significant correction. But going from that kind of armchair podcast statement to actually being able to make money on it — that’s damn hard.”

The counterpoint to all this short-selling cynicism? The revenue is actually showing up. OpenAI is projected to hit $20B ARR this year, while Anthropic projects $70B ARR by 2028. The growth rates aren’t just holding — they’re accelerating.

Harry drives this home: “Are we being overly British? Are we looking for a problem that’s not there? The revenue is showing up in billions.”

The lesson: It’s intellectually easy to be a skeptic about AI valuations. It’s financially hard to make money being that skeptic. And meanwhile, the actual revenue keeps proving the bulls right.

Gamma’s $100M Revenue Run Rate with 50 People: The New Efficiency Paradigm

Few stories capture the AI revenue opportunity better than Gamma’s latest round. The company raised $100M at a $2.1B valuation, having hit $100M in revenue with just 50 people and 2 million users.

Jason and Team SaaStr are super-users and he walks through exactly how they use it at SaaStr: “Instead of sending sponsors the same dated prospectus, Gamma automatically pulls all of our data from Salesforce and our marketing automation system. It knows the exact number of leads and ROI. It knows who their competitors are and makes a fully dynamic piece of collateral in about 10 minutes.”

Here’s the math that’s important: SaaStr pays $100/month for Gamma — $1,200 per year. That’s stealth TAM expansion. “How much do we spend for Google Slides? Zero. It’s built in. How much do we spend on PowerPoint? I don’t even know where my key is to Microsoft Office. So it is a stealth TAM expansion.”

But Jason’s most important insight is about what’s coming next: “The lame thing about co-pilots is they were just tools. When the AI is part of your team for real, not VC talk, the amount of revenue that it’s accessible is so high.”

He distinguishes between AI as a tool versus AI as a team member: “It is sufficiently autonomous, knowledgeable and powerful to complete material high-value tasks on its own with some daily discussions just like on our team. The level of autonomy and capability — Gamma, go into my Google Calendar, create prospectuses and sales collateral for all 20 sales calls this week, pull all the data from Salesforce and HubSpot and Marketo, review once, and distribute to the team.”

The implication: We’re transitioning from AI making us more efficient (2024) to AI being actual members of the team (2026). That’s where the revenue explosion happens.

But there’s a warning embedded in Gamma’s story too. Canva now has a Gamma clone that’s “pretty good.” The competitive response time has collapsed from years to months or even weeks. This leads to one of the episode’s most important debates…

Does Defensibility Exist Anymore? The Most Important Question in Venture

This might be the single most important strategic question facing every seed and Series A investor right now. And the panel doesn’t fully agree on the answer.

Jason frames the problem: “The quality of clones is only going up. I can think of one investment I’ve made that has had five clones in the first 30 days, including one from a cloud leader. The ability of AI to enable us to clone better stuff faster — Canva is borderline competitive with Gamma and wasn’t when Cliff was on the show. What does seed investing mean when you might see 10 better versions in 30 days?”

Rory’s response: “I don’t think you can have a major defensibility at the seed or even frankly the stage we’re investing at. The defensibility theorem emerges at scale. Once you become the anointed winner, once a market coalesces and there’s two or three people, you know, it’s yours to lose.”

But Rory pushes back on the idea that this makes seed investing impossible: “You just have to internalize the game you’re in. Awesome team, run fast, be superlative on technology, get your distribution early, and then rely on that. You can’t be anointed the winner up front. Get over it everybody.”

Jason’s concern is economic: “Is that okay at 50 post for a seed round? Do the outcomes justify it? If I’ve got to spread $5 million checks around at 50 post, it’s tougher.”

Harry asks the critical question: “Are we getting paid for the risk?” He notes that even at the Series B stage, it’s unclear who will win in many categories. Looking at customer support AI companies that have raised Series B rounds: “I got no idea who the winner is in that category and I don’t think anyone does to be honest.”

The debate crystallizes around whether you can know enough at the B to justify the valuation:

Rory’s view: By the Series B, you can see rank order. “Once the horses are running and once they round the first furlong, you can actually see the rank order of where they’re running in a way you can’t at the early A.”

Harry’s counter: Looking at codegen — even with clear leaders like Cursor emerging, you have Codex making incredible ground, Claude Code advancing, and Replit and Lovable coming from different angles. “I think that’s still entirely up for grabs.”

The practical implication: This uncertainty means one of two things:

* Accept that seed/Series A investing requires higher diversification (40+ companies instead of 20)

* Or accept lower ownership percentages betting on outcome expansion

Jason’s conclusion: “Maybe you need 40 deals at $5M. That’s $200M for first checks. $200M for reserves. That’s $400M. $100M for fees and backup. I need at least $500M for my little seed fund to make the math work.”

The Three Ways to Actually Capture AI Revenue

Amid all the complexity, the panel crystalizes a simple framework for what actually works in AI investing right now.

Path 1: Attach to Compute Infrastructure

DataDog crushed earnings, with stock up 23% and $15M+ in AI-native customers. Why? They’re selling to the people making AI, and as those companies grow, DataDog sells more too.

Jason: “The AI leaders, the hyperscalers — they’re starting to buy like classic B2B companies. They’re recycling the same people in procurement. So if you’re attached to AI budget and you’re a DataDog era, you’re actually going to have a great 2026.”

Rory adds: “DataDog is a core piece of compute infrastructure and these hyperscalers are the most compute-intensive companies that have ever been known. If you’re selling compute stuff, you should be having a great quarter. If you’re selling routers, switches, interconnects, whatever it takes to stand up Stargate — you’re going to be golden.”

Path 2: Replace Human Headcount

“Where are you replacing humans for real?” Jason asks. “Where are you going to go in and reduce the headcount that vendor needs by half?”

This is the Replit story — Jason deployed 20+ AI agents that generate significant revenue while requiring substantial daily management but far less human headcount than traditional approaches.

The key distinction: “Most of the time, ‘replacing humans’ is a story. But when it’s real, when the AI agent is sufficiently autonomous to complete material high-value tasks, that’s when you’re actually part of the team, not just a tool.”

Path 3: Massively Displace an Incumbent

“You have a third option,” Jason concedes. “Use AI to massively displace an incumbent and steal all the revenue. In fact, that’s the history of B2B software mostly. I just don’t know how many of our public leaders are in a place to steal their own revenue.”

But he’s skeptical this is as compelling as the first two: “There’s like 400 AI CRM startups out there all saying they’re going to eat HubSpot’s and Salesforce’s lunch. That’s not exciting to me as an investment. That’s much less exciting than truly replacing 90% of your GTM team.”

If you’re not doing one of these three things, you’re probably not going to see explosive growth. Using AI to make your product better? That’s table stakes now, not a competitive advantage.

How to Actually Run a Fundraising Process (Without Running a Process)

One of the most tactically useful segments came when Harry shared a founder interaction that frustrated him: the company said they wanted to “run a process” despite Harry offering terms they’d stated they wanted.

Harry’s reaction: “I said, listen, if you’re running a process, you’re either optimizing for price or partner selection. I’m giving you a great price today that you said you wanted, which means you’re not optimizing for price. Which is just saying you think you can get better than me. In which case, just tell me straight.”

Rory sees both sides: “From the founder’s side, they’re correct. I see more failed financings because they didn’t run a process than because they did.”

But then he makes a critical distinction: “What typically happens when people ‘don’t run a process’ is someone comes in, says they’re interested, and the founder shares information serially to someone not yet ready to commit. That’s a mistake. That’s running an accidental process.”

Jason has the most nuanced take on optimal strategy: “The best founders cultivate enough interest with enough good VCs that if they hit the number, they just send an email. Harry, I’m thinking about raising around before the end of the year. And Harry could say, I’ll give you a term sheet today. And the right answer is: ‘I love you, Harry. I’m not ready today. I will be ready at the end of the year.’”

The key insight: “The optimal way to run it is for everyone already to want to invest for real without games before you open your data room. The best run processes don’t require a data room. Not a traditional one. Only one for diligence.”

Rory synthesizes: “The best run processes don’t feel like a process, but they are. If a founder is smartly nurturing relationships, keeping people broadly informed, then tries to time the interest such that when they’re ready to put their hand up, there’s three people who are primed and ready to go — that is perfection itself.”

But here’s the brutal caveat: Jason points out that this only works if you have stellar numbers. “You’re either YC, Neo, South Park Commons, got something, get funded — or who the hell is going to find you? You better be hot AI-native with top-quartile venture growth or you ain’t getting funded.”

Harry shares a data point that proves this: A company that grew from $400K to $3M ARR (classic enterprise SaaS, 10x growth) had 120 meetings and got one term sheet at $10M on $40M post. That’s 12x revenue for a 10x grower — which five years ago would have generated five term sheets from top firms.

“It’s the most binary fundraising environment in our lifetimes,” Jason concludes. “You’re either Captain Obvious getting funded, or you’re not seeing it anywhere.”

Duolingo Down 25%: The Cautionary Tale of the “Wrong” Kind of AI

While DataDog surged 23%, Duolingo crashed 25% in the same week. The contrast illustrates everything about what’s working and what’s not in AI.

Jason’s diagnosis: “Duolingo has the ‘wrong’ kind of AI. Duolingo is using AI to make its product better. Hooray. Every single portfolio company at scale should be using AI by this point to make your product better. This is not 2023. You don’t get any kudos for sprinkling AI dust on your product.”

Rory pushes back slightly on behalf of consumer apps: “If you’re an infrastructure company, you can co-attach to compute. If you’re a new AI apps company, you are using that compute. But if you’re like Duolingo — preexisting — what you can do is co-adopt the technology.”

But Jason sees a bigger problem: “Duolingo took money from Berlitz and all these language schools. Hooray, you did that. Now, where are you going to disrupt humans? This is your job. You already disrupted those humans. Unfortunately, they’re gone. Where’s the next level of human disruption?”

Rory makes an interesting point about where the AI opportunity in education actually is: LLM learning is equivalent to one-on-one human tutoring. What AI should enable in education is allowing everyone to get one-on-one learning instead of group-based learning. But there’s no budget in K-12 to give every kid a customized tutor. The market is adults wanting to learn a second language for business who would pay for a one-on-one coach — a niche market.

Jason’s framework applies here too: “If you’re not removing humans from the equation, you’re going to be heavily discounted. You’re either getting money from compute or you’re getting money because you’re using AI to replace humans. Otherwise, congratulations on your 14% growth.”

Portfolio Construction: The Math Actually Matters

One exchange revealed how different the panel’s approaches are to portfolio construction — and why both can work.

Harry revealed his team’s meeting load: “With my four investing partners, we have 80 net new companies that we meet in person per week.”

Jason’s response: “I would resign. I would give you all my carry back. I sold my companies because I didn’t want to spend my life in meetings. I’ll do one a week, two max.”

But Jason’s critique cuts deeper than meeting volume: “When you do 200 deals between your team, you got to meet 500, 600, 700 founders a year. I don’t want to meet 500 founders a year. Unless you’re great, you’re going to blow my mind. But if you’re not, I’m gonna start yawning about 15 minutes into this meeting.”

Rory defends the high-volume approach: “Even on an okay deal, you learn nuances about a specific market from the one-on-one that you wouldn’t get from the presentation. The person inside living it every day has a crucial kernel of knowledge that you just can’t access any other way.”

But the conversation reveals a deeper truth about fund construction:

The Small Fund, High Multiple Approach (Hummingbird-style):

* Keep fund size at $40-100M

* Accept dilution in follow-on rounds

* Start with 20% ownership, end with 12%

* Generate 8-10x returns

The Large Fund, Maintained Ownership Approach (Lightspeed-style):

* Raise $250M+ funds

* Deploy significant follow-on capital to maintain ownership

* Generate 5x returns but on much larger capital base

Rory’s key insight: “Both are great outcomes. They’re just different ways to play the game. The interesting thing is both outcomes have great outcomes for the GP. But if you only have $1 to play with as an LP, you want the small fund 10x. If you have to deploy $100M, you have to do the big fund. The high small fund accept the follow-on dilution but make a marvelous return is the compelling product for the marginal dollar.”

The Hummingbird Outlier: How to Actually Generate 10x Returns

Speaking of Hummingbird, their Billion to One investment deserves its own section because it represents what’s possible when everything goes right.

Hummingbird did their first bio deal in Billion to One and now has an $800M position on the IPO — likely representing an 800x+ return on their initial investment from a sub-$100M fund.

Harry marvels: “You want freaking great venture returns? Credit due. Amazing.”

But Jason asks the tactical question every seed investor should be asking: “How did they collect the capital as a seed manager to deploy enough to maintain the ownership? I don’t see how I’ll ever own 18% of something at IPO ever again.”

The answers are telling:

* Capital efficiency — Billion to One didn’t require massive rounds

* Follow-on investment — They did put in subsequent checks

* Concentration — They bet big on the winners

* Accepting some dilution — Rory suggests they may have gone from 20% to 12% but on a massive outcome

Rory makes the crucial point: “You can accept some dilution and optimize for multiple rather than ownership. If you put in $4M and get diluted from 20% to 12% but it’s a $5B exit, you’re a hero.”

Jason counters with what’s exciting about capital efficient outcomes: “Navan at $4.5B, Billion to One at $5B — the power of capital efficiency and running lean delivers superior investor returns despite smaller headline valuations.”

The lesson: Small funds generating 10x+ returns by accepting dilution but betting on capital-efficient companies may be the optimal strategy for seed investors who don’t have the deployment capacity of mega-funds.

What This All Means: The Real State of Venture in 2025

Pulling all these threads together reveals several uncomfortable truths:

1. The Middle Class of Fundable Companies is Gone

“It’s the most binary fundraising environment in our lifetimes,” Jason states flatly. The classic SaaS triple-triple-double-double company? A company growing from $400K to $3M ARR would have had five term sheets from good firms. Now: 120 meetings, one term sheet.

2. Most People from the Last Decade Should Step Aside

This isn’t ageism — it’s pace of change. “The most folks from the last decade or 15 years are not the right people for the next decade,” Jason argues. “I struggle to even recommend a lot of the CROs and executives I know for roles today. We don’t even need half the VCs we have today for the AI world.”

3. Diversification Matters More, Not Less

With increased variance in outcomes and compressed time to clone, the math says diversification is critical. But diversification at seed prices requires massive funds or acceptance of lower ownership.

4. There Are Only Three Ways to Win

Attach to compute budgets, replace human headcount, or massively displace incumbents. If you’re not doing one of those three things, you’re probably building a slow-growth SaaS company in a world that doesn’t want those anymore.

5. Capital Efficiency Delivers Superior Returns

Billion to One at $5B and Navan at $4.5B generated better investor returns than most $20B+ outcomes because they raised less capital. The lesson: optimize for investor returns, not headline valuations.

6. The Best Fundraises Don’t Feel Like Processes

Cultivate relationships over months. Have 3-4 investors ready to commit before you formally raise. Send one email and get multiple term sheets. That’s the optimal path.

7. If You’re Not Excited, You Should Retire

Jason’s most important point: “If this isn’t the most exciting time of your lifetime, you’re doing it wrong. This is the first time software has gotten better since the three of us met. If you’re not incredibly excited, retire. No shame in that. Put the rest into NASDAQ — you’re going to make more than most VC funds anyway.”

Looking Forward: What 2026 Brings

The panel is clear about what’s coming: “This year was AI works. Next year is AI is part of your team.”

That transition — from AI as a tool making you more efficient to AI as an actual member of your team with autonomy, knowledge, and capability — is where the revenue explosion happens.

Jason’s final thought captures the optimism underlying all the tough talk: “If you’re not incredibly excited right now, if this isn’t one of the most exciting and stressful times simultaneously, you’re missing it. This is a great enormous mega trend. It’s the biggest mega trend maybe since the early days of the internet. Leaning into it is the only sensible thing to do, and playing against it is dumb as rocks.”

The old playbook is broken. The new playbook is being written in real-time. And the investors who figure it out — whether through small concentrated funds like Hummingbird or large diversified portfolios like Harry’s approach — will capture the extraordinary returns available in this moment.

Everyone else? Maybe it’s time to buy that beach house.

Quotable Moments

On Venture Evolution:

“The most folks from the last decade or 15 years are not the right people for the next decade. I genuinely believe it across my ecosystem. I struggle to even recommend a lot of the CROs and executives I know for roles today.” — Jason Lemkin

“The pace of evolution is so fast. If you decide what I knew 6 months ago is still useful, you’re probably going to be wrong very quickly. That’s what I find the most stressful about right now.” — Rory O’Driscoll

On AI Investment Strategy:

“Tools are great when the AI is part of your team for real, not VC talk. The amount of revenue that it’s accessible is so high.” — Jason Lemkin

“Sell to the people who are making AI and if they grow, you’ll sell more too.” — Jason Lemkin

On Defensibility:

“You just can’t take that early first month explosion as seriously as you used to. It’s not as defensible.” — Harry Stebbings

“I don’t think you can have a major defensibility at the seed or even frankly the stage we’re investing at. The defensibility theorem emerges at scale.” — Rory O’Driscoll

On Market Timing:

“It’s easy to be roughly right. It’s very hard to imagine that the AI capex boom doesn’t have a significant correction. But going from that kind of armchair podcast statement to actually being able to make money on it — that’s damn hard.” — Rory O’Driscoll

On Fundraising:

“The best run processes don’t require a data room. Not a traditional one. Only one for diligence.” — Jason Lemkin

“The best run processes don’t feel like a process, but they are.” — Rory O’Driscoll

On Current Reality:

“It’s the most binary fundraising environment in our lifetimes. You’re either getting funded or you ain’t seeing it anywhere.” — Jason Lemkin

On Taking Action:

“If this isn’t the most exciting time of your lifetime, you’re doing it wrong. If you’re not incredibly excited, retire. No shame in that. You had a great run. Put the rest into NASDAQ — you’re going to make more than most VC funds anyway.” — Jason Lemkin

Listen to the full episode of 20VC x SaaStr on YouTube or your favorite podcast platform. For more insights on venture capital, AI investing, and B2B metrics, visit SaaStr.com and subscribe to the SaaStr newsletter.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

By the end of Q3, we’ll had 10 distinct AI agents running in production at SaaStr. 20 if you including less critical ones. Not as a tech experiment or marketing stunt, but as core members of our revenue and operations team.

The lineup looks like this:

Revenue Team:

* 3 AI SDRs handling each of ticket inquiries, sponsor outreach, and sales support (these are different workflows, training, etc)

* 2 AI BDRs qualifying inbound leads and nurturing prospects through our funnel

* 1 AI RevOps agent tracking and managing our partner pipeline

Operations & Experience:

* 1 AI Support agent handling event logistics and attendee questions

* 1 AI Content Review agent vetting speakers and session proposals

* 1 AI Matchmaking agent connecting CEOs and executives at our events

Community & Education:

* 1 AI Mentor (SaaStr.ai) providing 24/7 guidance to our community. Try it, it’s free!

And we’re not done. The pipeline has 3-4 more AI agents in development.

The Operational Reality: It’s A LOT More Work Than You Think

Here’s what nobody tells you about AI agents in production: they require daily management and review. Not weekly check-ins. Not “set it and forget it” automation. Daily.

Every morning, I’m reviewing:

* Conversation quality scores from our AI SDRs

* Lead qualification accuracy from our BDRs

* Edge cases that required human escalation

* Performance metrics across all agents

* Training data updates and model refinements

Each agent needs constant fine-tuning. The AI SDR that handles sponsor inquiries needed 47 iterations to stop being too aggressive on pricing discussions. Our AI Support agent had to be retrained three times to properly escalate VIP attendee issues.

The truth? Managing 10 AI agents is like managing a team of 10 very capable but very literal junior employees who need explicit instructions for everything.

But Here’s Why We’re All-In: The Advantages Are Undeniable

Despite the management overhead, these AI agents deliver something human employees simply can’t:

They never quit. Zero turnover. No recruiting cycles. No onboarding new SDRs every 18 months because they got poached by a competitor offering $10K more.

They work weekends. While your human BDR is at Coachella, our AI BDR is qualifying leads and booking demos. Saturday morning inquiries get responded to in under 2 minutes, not Monday afternoon.

They don’t complain. No “this lead quality sucks” from the AI SDR. No “I need more training on the new product features” requests. They just execute.

They aren’t distracted. Our human SDRs were spending 30% of their time on side hustles, online courses, or job searching. The AI agents? 100% focused on converting prospects and supporting customers.

They scale instantly. Need to handle 3x more sponsor inquiries during SaaStr Annual planning? The AI RevOps agent doesn’t need additional headcount approval or three weeks of hiring. It just scales.

The Product Knowledge Advantage: They Know Everything “Cold”

This might be the biggest unexpected benefit: AI agents know our products, processes, and pricing cold.

Human SDRs need 3-6 months to really understand our event portfolio, sponsorship packages, and community offerings. Even then, they’re guessing on edge cases or scrambling to find answers. In fact, most of the SDRs we’ve had never really understood our products at all.

Our AI agents? They have perfect recall of:

* Every sponsorship package and pricing tier

* Historical attendee data and ROI metrics

* Speaker requirements and content guidelines

* Event logistics for 12+ annual events

* Community membership benefits and upgrade paths

When a prospect asks our AI BDR “What’s the difference between your Growth and Enterprise sponsorship packages for companies doing $50M ARR?”, it delivers a perfect answer in 30 seconds. No “let me check with my manager” or “I’ll get back to you”. Or worse, no making things up. (Our AIs are well enough trained that hallucinations are minor at best now.)

The Financial Reality: ROI Happens Faster Than Expected

The numbers are becoming undeniable:

Cost per agent: ~$200-4,000/month (including platform, training, and management overhead) Cost per human equivalent: ~$8,000-$12,000/month (salary, benefits, management, office space)

But the real ROI drivers:

* Response time: Average first response dropped from 4.2 hours to 1 minute

* Lead qualification: 67% more leads properly scored and routed

* Weekend coverage: 23% of our best leads come in outside business hours

* Consistency: Zero “bad days” or emotional decision-making affecting prospect experience

Our AI SDR team has generated $340K in sponsor pipeline so far in Q3 alone, and the quarter has just begun. At a fully-loaded cost of ~$10K/month for all core agents.

What Folks Get Wrong

“Mistake” #1: AI agents can’t replace human creativity and relationship-building. For complex enterprise deals and strategic partnerships, humans still close. But also, way too many in sales overestimate their skills here. Being a “people person” is not enough.

“Mistake” #2: Underestimating the management overhead. Many of you will need to hire a dedicated “AI Operations Manager” role to keep everything running smoothly. RevOps and MarketingOps are becoming radically different positions, requiring different skill sets, in the Age of AI. If you’ve had a bad experience with an AI SDR tool, it’s probably because you expected to buy-and-ignore it. Doesn’t work that way.

“Mistake” #3: Some prospects still prefer human interaction for high-value conversations. Although fewer than expected. Be transparent about AI involvement upfront. Many folks are happy to receive a >great< AI email that is truly hyper-personalized with >value<. With value hyper-personalized to their needs.

The Bottom Line for B2B Leaders

AI agents aren’t replacing your entire revenue team. But they’re becoming essential for:

* Top-of-funnel lead management

* 24/7 customer support and qualification

* Operational tasks that require perfect consistency

* Scaling during peak demand periods

The companies that figure this out in 2025 will have a massive operational advantage by 2026. The ones that wait for “better technology” or “clearer ROI” will be playing catch-up with teams that never sleep and never quit.

Our prediction: By SaaStr Annual 2026, the highest-performing SaaS companies will have AI agents handling 40-60% of initial prospect interactions. The question isn’t if this happens, but how quickly you can operationalize it without breaking your customer experience.

Start with one agent. Master the management overhead and the training first. Then scale.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

Medha Agarwal is a Partner at Defy VC, where she focuses on investments in AI-first and vertical SaaS companies. She shares insights at SaaStr AI Summit 2025 from the front lines of AI-first product pricing, exploring why traditional SaaS models are declining in favor of transactional pricing, how to choose the right pricing structure for your business, and strategies for capturing value from labor budgets instead of software budgets.

Top 5 Takeaways

* Transactional pricing is replacing traditional SaaS at an accelerating rate. The fundamental shift is driven by AI’s ability to complete tasks end-to-end, enabling companies to sell into labor budgets rather than software budgets. This opens up significantly larger TAMs that were previously dominated by human labor costs.

* There’s a 1.5-2.5x revenue multiple premium for SaaS models in public markets, but transactional models capture more value. While SaaS offers predictability and better cash conversion cycles with annual upfront payments, transactional pricing allows you to scale revenue with customer growth without being constrained by seat count. The trade-off is revenue predictability versus value capture.

* Hybrid models are emerging as the best of both worlds. Companies are mitigating transactional pricing unpredictability by implementing tiered subscriptions with usage minimums and overage billing. This provides baseline revenue predictability while maintaining the ability to capture value at scale through consumption-based pricing.

* Your pricing model choice depends on four critical factors. Frequency of usage, magnitude of cost savings, workflow integration point, and customer budget type all determine whether fixed-cost, input-based transactional, output-based transactional, or hybrid pricing makes sense. High-frequency tools like Slack need flat fees because users can’t mentally track per-use costs.

* Never compete on price alone when entering a market. Undercutting competitors on price creates a dangerous dynamic where you attract price-sensitive customers rather than best-fit customers, leading to higher churn and false signals of product-market fit. Price at par with competitors and win on value, or you’ll be forced into a race to the bottom.

The Fundamental Shift: Why SaaS Pricing Is Dying

We’re witnessing a massive sea change in how AI-first products are priced. At DEFY, we’ve seen a dramatic increase in the frequency of transactional-based pricing models. Traditional SaaS pricing, while still popular, is declining rapidly in favor of transaction-based approaches.

The reason is straightforward. With AI, there’s an increasing ability for software to complete tasks end-to-end. Some AI businesses are enhancing humans and making them significantly more productive than ever before. Others are eliminating the need for humans to perform many entry-level tasks altogether.

This creates a game-changing opportunity. It’s much easier today than it has been in the past to sell into labor budgets, which represent a much larger line item for most companies’ cost structures. Labor is also seen as more mission-critical than software spend. These transactional models can target much larger total addressable markets that were previously captured almost entirely by human labor expenditures, which are now starting to be displaced by software in certain categories.

The Three Pricing Model Archetypes

When we looked at the pricing landscape for AI-first products, we identified three high-level model types that most companies fall into.

Fixed Cost Models: Traditional SaaS

This is your typical traditional SaaS-based, seat-based pricing model. It can be seat-based or location-based pricing. There are compelling reasons why this model has dominated for so long.

If you look at the public markets, SaaS models command a 1.5 to 2.5 times revenue premium multiple compared to transactional-based models. There are several reasons for this valuation gap.

First, it’s extremely predictable for the company selling the software. Second, it’s very predictable for the customer buying it. They can plan for that spend on an annual basis with high confidence. Third, the cash conversion cycle is excellent. These contracts are often paid yearly upfront, which helps fast-growing companies manage their cash flow effectively.

But there are significant trade-offs with fixed pricing models.

As companies grow rapidly, their usage of a product often scales much more quickly than the number of seats they need. This happens because they’re not growing headcount proportionally since the tool is making them much more efficient. The result is that the value a vendor is able to capture is much less than the actual value they’re providing to the company.

Second, there are use cases where you could add enormous value, but the usage isn’t consistent on an everyday basis. This makes SaaS pricing less sensible. Financial planning tools are a perfect example. There will be intense periods of usage, followed by periods of much lighter usage. Paying for a full-time seat year-round doesn’t align with the value delivery pattern.

Transactional Pricing: Input-Based vs. Output-Based

When we think about transactional pricing, it breaks down into two distinct categories: input-based and output-based.

Input-based pricing is when you bill on consumption or usage that ties very closely to traditional API-based usage pricing. You charge for the number of pages processed, the amount of compute consumed, or the number of API calls executed. This model is familiar to developers and technical buyers who understand the cost structure underneath.

Output-based pricing is fundamentally different. It’s much more aligned with how humans work and what they deliver. What is the output? What is the unit of work being completed? If you successfully complete a task, you bill for that unit of work or that task. This is closer to how you’d pay a human contractor or employee for deliverables rather than for effort.

The benefit of transactional-based pricing, whether input-based or output-based, is that you can charge more for each unit of work you’re performing. You can scale revenue much more efficiently with the growth and expanding usage of your existing customers. As they do more, you earn more, without negotiating new contracts or adding seats.

The downside is predictability. You can’t forecast revenue as confidently. There might be a quarter where your revenue is off the charts because customers heavily used your solution. Then another quarter comes where they don’t need to use it as intensively, and revenue drops. This makes it much harder to predict revenue and model your business for investors and your board.

The Hybrid Model: Having Your Cake and Eating It Too

We’ve seen many companies successfully mitigate transactional pricing unpredictability by implementing hybrid models. These combine elements of fixed tiered base subscriptions with usage minimums and consumption-based billing.

Here’s how it typically works. There are three tiers. For tier one, you pay X amount on a monthly or annual basis as a minimum commitment. But that subscription includes X number of credits, either input-based or output-based. If you use more than your included credits, you move into the next tier, and the next tier after that. Companies then back-bill for overage at the end of every quarter based on actual usage.

This structure provides baseline revenue predictability while maintaining the ability to capture value as customer usage scales. It’s becoming increasingly popular among AI-first companies that want to balance investor expectations for predictable revenue with the reality that AI products create value through consumption.

The Evolution: Aligning Price with the Value of Labor

AI-first products, enabled by the evolution of technology, are making pricing models possible that align much more closely with the value of labor rather than the value of software. This is something we’re really excited about at DEFY.

We’ve seen tremendous potential for companies to build very large businesses on top of this ability to target labor spend. The opportunity is massive because labor budgets dwarf software budgets in almost every organization. If you can credibly replace or augment labor with software, you can charge proportionally to the labor cost saved, not just the software value delivered.

How to Choose Your Pricing Model: The Four Critical Factors

It’s extremely important to think through the specific inputs and your unique situation when determining which pricing structure makes the most sense for your business. Every model has trade-offs. Here are the four critical factors you need to evaluate.

Factor 1: Frequency of Usage

How often are customers using your tool? Is it something they use day in and day out, many times per day? If so, usage-based pricing could be really complicated and might gate adoption because users would think twice before using the solution.

Slack is the perfect example. If Slack used usage-based pricing, it would be much harder for you or me to keep track of whether sending a Slack message was worth X cents. We’d mentally calculate the cost before every message, which would create massive friction and reduce usage. That’s a type of product where the frequency of use is so high that customers can’t reasonably predict their costs. A flat fee makes much more sense.

Conversely, if your tool is used infrequently or sporadically, usage-based pricing makes perfect sense. Customers only pay when they get value, and they can clearly see the ROI of each interaction.

Really think carefully about what the impact of pricing either on a fixed basis or usage basis will have on customer adoption and expansion. High-frequency use cases almost always favor fixed pricing to remove mental transaction costs.

Factor 2: Magnitude of Cost Savings

The second factor is understanding how much you’re saving the customer compared to the alternative, which is often human labor. If you’re delivering a 10x cost reduction compared to hiring someone to do the work, you have much more pricing power and can justify premium transactional pricing.

For example, if a customer would need to pay $50 per hour for a human to complete a task, and your AI can do it for $5 worth of usage, that’s a compelling value proposition that justifies transaction-based pricing. You’re directly comparable to labor costs, and the customer can easily calculate their ROI.

But if your cost savings are more marginal, or if the comparison to labor is less direct, you might struggle to justify high transaction prices. In those cases, fixed pricing might make more sense because it’s harder for customers to calculate and justify the per-use ROI.

Factor 3: Integration Point in the Workflow

Where you insert yourself into the customer’s workflow dramatically impacts your pricing model and revenue predictability.

Let’s walk through an example. Imagine you’re building a customer service AI solution. If the customer wants to route all of their incoming support volume through your solution first, and you handle everything you can before escalating to humans, you have massive predictability. You can look at their historical support ticket data and have a pretty good sense of the volume you’ll process. You can plan accordingly, and they have predictability in their costs.

But if they only want to insert you somewhere later in the funnel after a human has already triaged the tickets, and they only route the most complex queries to you, it’s much harder to predict volume. They’re deciding when to bring you into their funnel and what percentage of volume to send your way. Your revenue becomes dependent on their operational decisions, which introduces significant unpredictability.

If you can work with customers consultatively and explain why you should be at the top of the funnel processing all volume, you can help them understand why it’s beneficial to them while also giving yourself more predictable revenue. We have a portfolio company working with logistics companies where they’re the first line of contact when anyone calls. They answer what they can in an automated way, and only route complex issues to humans.

Their argument is compelling. The vast majority of customer needs when they call can be automated. By handling those, you’re reducing call wait times, which leads to lower drop-off and probably reduces lost revenue. Your human team can focus on high-value, complicated tasks instead of spending expensive human time on entry-level support tickets. They charge based on the number of minutes or calls or tickets they resolve, and the customer can predict their costs based on total incoming volume.

Factor 4: Budget Type and Buyer Persona

Understanding whether you’re selling into a software budget or a labor budget fundamentally changes your pricing model and your competitive landscape.

If you’re truly replacing labor, you’re competing against fully loaded employee costs, including salary, benefits, training, management overhead, and retention risk. That’s a much larger budget pool and allows for much higher pricing. You can justify transactional pricing based on tasks completed because you’re directly comparable to what they’d pay a human.

If you’re selling into a software budget, you’re competing against other SaaS tools for limited IT spend. This typically favors fixed, predictable pricing models that IT buyers can budget for annually. The comparison is software versus software, not software versus people.

The buyer persona also matters enormously. A finance or operations leader buying a labor replacement is thinking in very different terms than an IT leader buying software. The finance buyer thinks in terms of cost per unit of work, productivity gains, and headcount planning. The IT buyer thinks in terms of licenses, integrations, and annual contracts.

The Pricing Strategy Question: Never Compete on Price Alone

One of the most important strategic questions early-stage companies face is how to price relative to competitors. Should you charge premium prices, match the market, or undercut competitors to gain initial traction?

I’m not a huge fan of trying to enter a market and take share by underpricing your competitors. That’s a slippery slope for several reasons.

First, if you win on price, competitors can also drop their prices. You haven’t built any sustainable competitive advantage. You’re in a race to the bottom where everyone’s margins compress.

Second, you attract the wrong customers. Are they signing up with you because you’re the cheapest option, or because you’re the best fit for what they need? Price-sensitive customers are often the worst customers. They’ll churn as soon as someone cheaper comes along, or they’ll constantly negotiate for discounts.

Third, it creates false signals of product-market fit. You might see strong initial adoption and think you’ve nailed product-market fit, when in reality you’ve just found a group of bargain hunters. When you try to raise prices or competitors match your pricing, you’ll discover that the demand was price-driven, not value-driven.

That means you might have higher churn down the road, and customers might demand you drop prices further when your competitors inevitably respond. There’s a real risk to scaling your business thinking you have product-market fit when you actually don’t.

I totally agree with founders who want to price at par with existing competitors rather than undercutting them. If you can’t win at comparable prices, you probably don’t have a strong enough value proposition yet. Fix the product and the positioning before trying to fix the problem with pricing discounts.

Determining the Right Model for Your Specific Business

Whether to charge for usage or flat fee or seat-based pricing requires deep customer discovery. You need to understand how customers see your product and the value it delivers.

Do they see it as a labor replacement? If you’re able to do something end-to-end or truly improve the productivity of an individual, you can make a compelling ROI argument for why you should charge a premium price on a usage basis for that work.

Or if they’re using it so much and so frequently that usage tracking becomes burdensome, you need to recognize that. If they’re only going to use your product for the most complicated, edge-case scenarios, then you’re actually limiting your scope and your revenue potential. You might be better off with flat pricing that encourages them to send you more volume, including the easier cases that build your data flywheel and improve your models.

Understanding what your customer is optimizing for is crucial. Are they optimizing for cost predictability? For variable costs that scale with their business? For ROI on specific tasks? The answer should drive your pricing model.

Your First Customers Matter More Than You Think

Being very careful about who your first five to ten customers are is critically important for AI-first products.

You likely have to do significant work, customization, and training to make your product effective for each early customer. If they’re not willing to pay you properly, or you don’t have confidence that customer is going to stick around and provide valuable feedback, it might not be worth the effort to sign them up at all.

Early customers set the precedent for your pricing, your positioning, and your product development roadmap. If you start with customers who are only willing to pay bargain-basement prices, you’ll struggle to raise prices later. If you start with customers who see you as a nice-to-have rather than a must-have, you’ll build features that serve that perception.

Focus on customers who have the problem you solve most acutely, who have budget to pay for solutions, and who will be true partners in developing the product. Don’t optimize for customer count in your first ten or twenty deals. Optimize for customer quality, reference-ability, and learning.

The Bottom Line

Pricing AI-first products is fundamentally different from pricing traditional SaaS because the value proposition is different. When you can complete tasks end-to-end or dramatically amplify human productivity, you’re competing with labor costs, not software costs. That opens up massive TAMs and justifies transactional pricing models that scale with value delivered.

The key is understanding your specific situation across the four critical factors: usage frequency, cost savings magnitude, workflow integration point, and budget type. There’s no one-size-fits-all answer. SaaS pricing still makes sense for high-frequency tools where usage-based billing would create friction. Transactional pricing makes sense when you’re directly replacing human work or processing variable workloads. Hybrid models are emerging as a powerful middle ground that provides predictability while capturing scale.

Whatever you choose, don’t compete on price alone. If you can’t win at market rates, you don’t have product-market fit yet. Fix that first. Your early customers and your early pricing set the trajectory for your entire business. Get them right, and everything else gets easier.

The era of pure SaaS pricing is ending. The companies that figure out how to align their pricing with the labor value they deliver, rather than the software value they provide, will capture the largest TAMs and build the most valuable businesses in the AI era.

*

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

So we’ve spent a ton of time over the years on SaaS talking about hiring a great VP of Sales / CRO . Not only because it really matters, but because hiring the wrong VP of Sales can set you back a year — or longer.

So I thought I’d come back to the classic topic and make a list of the Top 10 Mistakes I See Founders Make When Hiring a VP of Sales:

#1. Hiring a VP of Sales Who Never Really Understands Your Product During The Interview Process

Ok I know some even many will disagree, but I’m right here :). I can tell you as a pretty good investor across many leading B2B companies, I’ve never seen a VP of Sales thrive that didn’t really understand the product during the interviewing process. Never. I see so many B2B startups hire someone likeable, who can talk the talk on sales hiring and processes — but never really understands what you do. Or puts in the effort to do so. Don’t make this hire. They never invest the time after they start, either. Or they are never able to.

This has almost become my #1 flag now.

Way too many folks give managers a pass here that never understand the product. You gotta watch the YouTube videos. Do a demo. Listen to some Gong calls. At least get close. Or you just plain never do once you start. So many VPs of Sales disagree with me here — at least at first when I make the point. But later, they agree 😉

#2. Hiring a VP of Sales With No One Lined Up to Follow Them

This is a classic SaaStr point and post from over the years, and it turns out it’s more true today than ever. 50% of what a VP of Sales really does is recruiting. So the best VPs of Sales always have at least 2-3 great folks lined up to come with their to their next role. Just ask. Ask who those 2-3 are. And if you’re ready to extend an offer, talk to them before you do.

#3. Hiring a VP of Sales That Actually Doesn’t Want to Sell Themselves Anymore

This one has really become an issue in recent years, and the one hand I get it. Sales is hard. And it never really gets easier. So at some point in their careers, some some leaders don’t really want to sell themselves anymore. They’ll manage a team. Check the dashboards. Build process. But sell themselves? They’re sort of done. We call this Mr/Ms. Dashboards, and it’s not a new thing per se. But it’s much more common than a few years back. Because SaaS is getting to be 20+ years old.

Don’t hire this person. No matter how well they can talk the talk.

#4. Hiring a VP of Sales That Doesn’t Want to Go Visit Customers In Person

This is newer, but common these days. I recently interviewed a seasoned VP of Sales that lived in the South Bay in the Bay Area. He said he wouldn’t travel all the way to SF to visit customers because it was “too far”. I get it, with traffic, it can take 90 minute. But give me a break.

There are sales jobs that are 100% on Zoom. But you gotta at least visit the bigger ones. Many don’t want to do that anymore after years of working from home. Unless you sell 100% to SMBs, probably don’t make this hire. Ask.

#5. Hiring a VP of Sales That Doesn’t Want to Close At Least Some Customers Themselves

Your VP of Sales can’t carry a bag forever, at least not a full quota. But I’ve come to see that a new VP of Sales that doesn’t want to close deals themselves when they start often never really learns how to do it at all. A VP of Sales candidate that insists on closing deals themselves when they start? A great sign. One that says it doesn’t matter, that it’s all process? Maybe run.

#6. Hiring a VP of Sales That Has Gotten Cyncial on Startups, Tech, and Sales

Something I didn’t use to see much, but now is pretty common. I get that everyone has a tough startup experience or two. But if you can’t get past it, if the “system is rigged” against you … well I hear you. But don’t make this hire. You need Pirates and romantics in a startup, folks whose energy drives and guides and leads the team. Not someone who sees the whole system rigged against them.

#7. Hiring a VP of Sales Constantly On Social Media, Especially LinkedIn

I do believe some of this promotion is good. It helps with recruiting, and more. But the VPs of Sales that are posting 2-3 times a day on LinkedIn? I’ve found they really want to be influencers, advisors, etc. They don’t really want to do the tough, full-time job of VP of Sales. I know some will challenge me here. A few great posts a week on social can be good. But a few a day? Run.

And yes, I know I and SaaStr post a lot on the socials 🙂 But that’s our job, folks.

#8. Hiring a VP of Sales That Really Wants to Be COO, CRO, etc. And Not Really Be a VP of Sales.

Don’t force someone here. If a VP of Sales is done with that role and really wants a “bigger” job where they don’t just own the new bookings number, that can have a place. But it’s not as your VP of Sales. Now a little titlle inflation IMHO isn’t the end of the world. If your VP of Sales wants to be called CRO but their real job is VP of Sales, not also owning marketing, customer success, etc. — that can be OK. As long as you’re 100% clear here. 100% clear.

#9. Hiring a VP of Sales That Hasn’t Been a VP of Sales in a While But Wants to “Get Back to Sales”

I get this might work in a few cases. But 95 times out of 100, don’t make this hire. VP of Sales is a tough job. Taking a short break? No problem. But going off and doing something else for a long time? I rarely see them really able to get back in the saddle again. Once in a while, yes. But understand it’s a big risk you are taking.

#10. Hiring a VP of Sales That You Wouldn’t Hire If They Hadn’t Worked At ________.

Ok my top flag is where we started the post — a VP of Sales that never understands what you do. But this one is close. Everyone gets blinded by that great LinkedIn, by that fancy logo on the resume. If you love Doug, but in part it’s because he worked at Datadog, or Snowflake, or Asana, or wherever, so be it. Just be honest. Would you still hire him or her if they hadn’t worked there?

I ask founders to do this one exercise: block that fancy logo off from their LinkedIn. Literally, with your hand. Now, would you still hire them? If so, go for it. If not? You’re being blinded.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

Brett Queener is Partner at Bonfire Ventures, a $1B AUM seed-stage fund writing $3-4M checks into application software companies. He was employee #70 at Salesforce.com, where he built go-to-market, launched the AppExchange, and helped scale the company from its earliest days. Previously, he worked at Siebel Systems (the fastest-growing software company of its era) and ran a B2B startup (SmartRecruiters) from pre-revenue to $100M ARR. He writes about the changing software industry in real-time at his Substack.

He came to SaaStr Annual + AI Summit for a deep dive on AI and Product.

Brett’s Top 5 Take-Aways

* Your product has to deliver immediate, “What The Frack” Value Now in the Age of AI. It has to immediately do a job you couldn’t do before.

* Start small, expand fast. Forget the big enterprise land. Demo with their data, put it in their hands immediately, let them feel the “holy s**t” moment with an agentic assistant—then expand. The McGillaguer-Guerrilla deal is over.

* Your product must teach itself. When you’re shipping every 30 days, quarterly release webinars are dead. Build agentic assistants that tell users: “Hey, you know I can also do this? Want to try it?” The product needs a relationship with the user.

* Rethink annual contracts. If agents behave like $200K employees (paid monthly, can be fired), why are we doing annual upfront payments? The renewal decision isn’t “does our software need to keep running”—it’s “is this assistant still the best person for this job?”

* Fire customers who don’t get it. Some enterprise buyers want 12-month Accenture rollouts. They’re treating your agentic solution like it’s PeopleSoft in 2003. Walk away. They’ll slow you down until you die.

What happens when product innovation accelerates 10x? Everything you know about building SaaS is about to change.

I’ve been in enterprise software since the green screen days. Built CRM systems on Access and Visual Basic. Was employee 70 at Salesforce when we spent 60 cents of every dollar on our own data centers. Ran a startup that hit $100 million ARR (through a lot of tears, I’ll admit). Now I’m a partner at Bonfire Ventures, writing $3-4 million seed checks into application software companies doing $500K-$1M in revenue.

And I’m anxious. Not the normal founder anxiety. A different kind. The kind that comes from watching the fundamental rules of software change in real-time.

The Old Playbook is Dead

When SaaStr started, the model was simple: build a SaaS version of an on-premise category winner. Ship a big product release once a year at Dreamforce or your equivalent user conference. Run the company full-throttle across the entire organization based on that annual cadence.

Every blog post, every playbook, every piece of advice you’ve read about scaling SaaS? It’s all predicated on a world where your product changes once a year.

But what happens when your product has to change every 30 days?

Because if you’re a founder in this room right now, that’s your new reality.

We’ve Never Seen This Pace Before

I’m 55 years old. I was the first person at my company to buy a PC. I built CRM systems by carving regional manager data off IDEO drives and FedExing hard drives around the country. There was no email.

I’ve watched the internet emerge. Watched SaaS kill on-premise. Watched mobile kill desktop. Watched cloud infrastructure commoditize.

And I’m telling you: we have never, ever seen this pace of change in technology.

Not even close.

The Agentic Revolution Changes Everything

Here’s what’s different about AI agents versus every other technology shift:

When you give someone an agentic assistant that actually helps them do their job for the first time, their reaction is: “Holy s**t.”

Not “this is interesting.” Not “I’ll think about it.”

Holy s**t.

That’s a different buying motion. That’s a different renewal decision. That’s a different everything.

The Three Fundamental Shifts You Need to Understand

1. The Job-to-be-Done Framework Just Got Complicated

Think about pricing for agents. Here’s my framework:

* If AI helps the existing user do their current job better: Include it in the base price. Don’t charge extra.

* If AI solves an additional job for the same user: Consider charging for it.

* If AI does a brand new job for a new user in the org: You can charge for it.

I haven’t fully wrestled this through. Nobody has. Manny (who just left Outreach) is building a CPQ platform specifically for agent pricing. That’s how nascent this is.

2. Your Sales Motion Must Be “Start Small”

I have a three-time successful founder in my portfolio right now. Tried to sell a full enterprise solution into a vertical with legacy, non-cloud systems. The buyer came back and said Accenture would do a 12-month rollout, starting in Ulaanbaatar before tier-one countries.

I told him: “Go tell the customer they’re f*****g stupid. Tell them they don’t get it. We’re not doing 12-month rollouts.”

The new motion:

* Demo with sample data

* Demo with their data

* Put it in their hands immediately

Whether that’s self-serve, paid trial, or assisted doesn’t matter. What matters is they start using it and feel that “holy s**t” moment.

Then you expand.

3. The Product Must Teach Itself

When you’re shipping major updates every 30 days, the old “quarterly release webinar” model is dead.

The future? Your agentic assistant says: “Hey Brett, you’re using me for X, Y, and Z. You know I can also do A? Would you like to try it?”

The product has to have a relationship with the user that helps them discover new capabilities as they ship.

What This Means for Your Company Structure

If you’re shipping quarterly in 2025, you’re already behind.

But here’s what nobody’s talking about: when you move to continuous shipping, your entire go-to-market org has to change with it.

I used to invest in SaaS companies that said “we ship daily.” I’d tell them: “I promise you’ll move to quarterly releases. Once you build a sales team, success team, and customer team, you’ll need coordination.”

That’s still true. But the coordination has to happen at 10x speed.

The challenge: How do you ensure your customer-facing teams know about new capabilities fast enough to drive adoption before the next release ships?

Most companies will fail at this. Their product will race ahead while their GTM team sells last quarter’s product.

The Renewal Decision is Fundamentally Different

When I hire a $200K assistant, do I pay them $200K upfront? No. I pay them monthly. Can I fire them? Yes (with some severance).

So why are we still doing annual upfront payments for agentic solutions that behave like employees, not software?

The renewal decision for an agent isn’t: “Does our Salesforce org need to keep running?”

It’s: “Is this assistant still the best person for this job compared to other assistants I could hire?”

That changes everything about retention.

Your customers need to understand all your capabilities in a way they can consume. Not once a quarter. Not through a CS quarterly business review. Continuously. In the product.

The Uncomfortable Truth About Enterprise Buyers

Some enterprise buyers just suck.

They don’t understand this shift. They’re still thinking in 12-month implementation cycles. They’re routing everything through Accenture.

And here’s what I’m telling my founders: You might need to fire those customers.

Not literally (usually). But you need to be willing to walk away from buyers who want to treat your agentic solution like it’s PeopleSoft in 2003.

Because if you let them slow you down to their pace, you’ll die. Some competitor who’s shipping every 30 days will eat your lunch.

What To Do Right Now

If you’re a founder reading this, here’s your checklist:

Immediate (This Week):

* Audit your product release cycle. If it’s longer than 30 days, you have a problem.

* Look at your last 5 deals. How long from demo to “hands on keyboard”? If it’s more than 2 weeks, you have a problem.

* Check your pricing. Are you charging for AI capabilities that make the core job better? Stop.

This Quarter:

* Build the mechanism for your product to teach users about new capabilities

* Restructure your sales comp to reward “start small, expand fast” not “land the biggest deal possible”

* Identify which enterprise buyers “get it” and which don’t. Plan accordingly.

This Year:

* Question everything about annual contracts and upfront payments

* Build the muscle to ship major capabilities monthly, not quarterly

* Accept that your GTM playbook from 2019 is dead

The Bottom Line

We’re in the change economy now. The rules that governed SaaS for the last 20 years are being rewritten in real-time.

Product innovation is 10x faster. Buyer expectations are fundamentally different. The very nature of what “software” means is shifting from tools to teammates.

Most companies won’t make the transition. They’ll keep running the old playbook, shipping quarterly, doing 12-month enterprise rollouts, charging annual upfront.

And they’ll die.

The companies that survive will be the ones who embrace this pace of change. Who ship every 30 days. Who start small and expand fast. Who let their product teach itself to users.

It’s uncomfortable. It’s anxiety-inducing. We’ve never seen anything like this before.

But that’s exactly why it’s the biggest opportunity in software since the shift from on-premise to cloud.

Are you moving fast enough?

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

A SaaStr Annual + AI Summit conversation with Jesse Zhang, CEO of Decagon, and Sarah Ittelson, Partner at Accel

Decagon Today: The Numbers Behind the Hype

Founded in late 2023—just months after GPT-4’s release—Decagon has become one of the fastest-growing AI companies in history. The company builds AI customer service agents for large enterprises, automating conversations that previously required human support teams.

The Growth Trajectory:

* Founded: Late 2023

* Time to Eight Figures ARR: ~18 months

* Team Size: ~100 people (and scaling rapidly)

* Location: 100% in-person team

* Customers: Major enterprises including Hertz, Chime, and other leading brands

* Typical Customer ROI: $800K in savings for every $250K spent

* Market Position: Recognized as the leading Gen-AI native solution in customer service automation

What Makes This Growth Unprecedented:

Even by venture standards, this is exceptional. Sarah Ittelson, the Accel partner who led their Series A investment, has been part of the hyper-growth phases at Uber, Uber Eats, and Fair. Her assessment? “This current moment and the scaling that’s possible within these AI companies is unparalleled to even those hyper-growth moments of before.”

When Accel invested at the Series A, Decagon was targeting seven figures. By the time Jesse and Sarah took the stage at SaaStr to share their playbook, they’d already blown past eight figures. The headline had to be updated mid-flight.

This isn’t a story about getting lucky in a hot market. It’s a masterclass in intentional decision-making, relentless customer focus, and building a machine that compounds growth. Let’s unpack exactly how they did it.

The Market Selection Framework: Why Customer Service Won

Here’s the reality most founders miss: your growth rate is mostly determined by which market you’re in.

Jesse and his co-founder didn’t just pick customer service because it seemed like a good idea. They ran a rigorous discovery process—talking to roughly 100 potential customers over the course of a month. Every day packed with customer conversations. Every night cranking out product to show the next day.

What made customer service the winner?

* Clear, measurable ROI: Companies could point to specific dollar savings. Spend $250K, save $800K in human support costs. That’s not a pitch—that’s math.

* Massive TAM: Customer service is one of those rare markets where the surface area is enormous. Every large company has support teams. Every user interaction is a potential automation opportunity.

* Buyer urgency: Unlike many SaaS categories, companies were willing to move off schedule to adopt AI solutions. The business case was too compelling to wait for the next budget cycle.

The key insight: they weren’t looking for a market where AI could work. They were looking for a market where companies were already bleeding money on a problem that AI could solve today.

The Customer Discovery Process: 100 Conversations Before a Single Line of Code

Let’s get tactical about how Decagon approached early customer discovery, because this is where most founders either win or waste months.

The Daily Cadence:

* Pack every day with as many customer conversations as possible

* Extract commitments: “If we built this, how much would it be worth to you?”

* Build at night based on what you learned during the day

* Show it to customers the next day

* Iterate ruthlessly

What They Were Really Listening For:

Not excitement. Not validation. They were listening for willingness to pay.

Here’s Jesse’s framework: You could talk to the wrong person who feeds you useless feedback. Or you could talk to someone who’s super excited about your pitch, but when you get to pricing, there’s nothing there. True discovery means being aggressive about understanding what customers actually value and what they’ll actually pay for.

The Anti-Pattern They Avoided:

Jesse’s previous company was a consumer startup (eventually acquired by Niantic). The biggest lesson? They spent too much time sitting by themselves thinking about good ideas, building them, launching, and getting zero traction. That’s the burnout loop.

With Decagon, they flipped the script: talk first, build second, validate constantly.

Staying Close to Customers at Scale

Here’s where it gets interesting. Most companies do customer discovery well early, then lose touch as they scale. Decagon hasn’t.

How they maintain customer intimacy:

* Weekly touchpoints with every customer (they work only with larger customers, which makes this feasible)

* Proactive outreach—don’t wait for customers to complain

* The entire team stays involved in go-to-market, not just sales

The principle: customers might not proactively tell you what’s missing. You have to pull it out of them. You have to understand what’s next on the horizon before they fully articulate it.

Building the Team: Why In-Person and Intelligence Matter

When Sarah Iden first met with Decagon, Jesse was closing employee number five. Even at that stage, the focus on talent was exceptional.

The Decagon Team Philosophy:

They’re essentially 100% in-person. That’s not an accident—it’s a forcing function for the culture they wanted to build.

What They Screen For:

* Raw intelligence and adaptability: At the early stage, there’s no structure. No playbooks. No one hands you a to-do list. You need people who can figure it out.

* Commitment level: This is the non-negotiable with co-founders and early employees. There’s a wide range in how committed people are to being a founder or early-stage employee. If it’s not aligned, nothing else matters.

* Working style alignment: Are people working at the same pace? Putting in similar time? Gearing toward the same goals?

The Interview Process:

They ask about past experiences: “Tell me about a project where you invested the most time in your life.” They look at favorite projects, how hard people worked, the pace they maintained. They get back-channel references. By the time someone reaches the final interview with Jesse, they’ve already opted into the in-person culture and the intensity.

There’s no hard sell needed. The screening has done its job.

The Co-Founder Equation: Finding Your Match

Jesse’s co-founder relationship formed through A16Z (who backed his previous company) and mutual friends. But the chemistry wasn’t luck—it was alignment on fundamentals.

What Must Align:

* Similar life stage

* Level of commitment (the absolute non-negotiable)

* Working style and pace

* Type of company you want to build

* Amount of time you’re willing to invest

What Can Differ:

* Specific skill sets

* Personality traits (to a degree)

The Best Way to Find Out:

Build something together. Do a trial. If you’re considering co-founding, you probably have the time to experiment. This is unique to co-founder dating—you can’t get this kind of trial when hiring someone with a full-time job.

The trial will surface everything: working style, commitment level, how you handle disagreement, pace of execution. Everything else is just conversation.

The Execution Culture: How to Maintain Intensity at 100 People

In the early days, execution culture is easy. Everyone’s in one room. You’re talking constantly. Alignment is organic.

At 100 people, it requires systems.

Decagon’s North Star Approach:

Everyone needs to know what they’re working toward. If you’re in sales or engineering, you need clear goals aligned in the same direction. If you’re not working toward one of the important goals, you’re probably not working on something important.

The Metrics That Matter:

For a SaaS company, it’s simple:

* ARR

* Customer count

* Customer quality

These are measurable. They become the north star everyone marches toward.

The Celebration Cadence:

They celebrate wins constantly:

* Customer gives great feedback

* Customer goes live

* Deal closes

* Customer renews for two years

Every win gets shared. The team knows what they’re working toward and they see the impact in real-time.

The Challenge Ahead:

Jesse’s candid about this: they’re still figuring out how to maintain this culture as they scale. Right now they’re about 100 people and growing quickly. At some point, you can’t maintain culture just by talking. You need systems. They’re actively building those now.

The AI-Native SaaS Motion: Why Everything’s Different

Here’s what’s fundamentally changed about selling SaaS in the AI era:

1. ROI Must Be Provable and Fast

You can’t hand-wave anymore. Customers expect to test and see real ROI. For Decagon, that means:

* Quantifiable conversation resolution rates

* Measurable cost savings

* Improved customer satisfaction scores

If both cost savings and CSAT are going up, the business case sells itself. They’re saving money AND customers are happier. That drives future revenue and retention.

2. Deployment Can Be Incremental

You don’t have to deploy all at once. Pick one surface area. Deploy to 10% of your customer base. Get live in production. If it works, ramp up.

This makes buying cycles shorter and happens more frequently because you’re not waiting for massive projects with perfect test cases before commercial discussions begin.

3. The Technology Layer Keeps Changing

Models are improving constantly. New techniques emerge. You have to build with the assumption that everything underneath will change.

How Decagon Handles This:

* Build for model flexibility—instant evaluation and swapping of new models

* Own the application layer (the final product customers buy)

* Focus on solving the business problem, not showcasing the technology

The business problem doesn’t change even as technology evolves. That’s the power of building at the application layer. You’re evaluated on solving the problem, and as technology improves, your product improves automatically.

Breaking Through the Noise: Product Marketing in a Hyped Category

Every company is being told to have their “AI narrative.” So how does Decagon cut through?

Message #1: Gen-AI Native

Customer service has a long history of automation. The natural question is: “How are you different from existing solutions that are now adding Gen-AI?”

Decagon’s answer: They built from the ground up for Gen-AI. The entire paradigm of how you build these agents is different. You can do fundamentally more when you’re not retrofitting Gen-AI onto legacy architecture.

Message #2: Real ROI with Real Customers

AI is hyped. But Decagon has case studies with companies like Hertz and Chime showing:

* Specific dollar savings

* Support team repurposing

* Higher NPS and retention

The message isn’t “AI is cool.” It’s “AI actually works here, and here’s the proof.”

Goal Setting: Maintaining Healthy Dissatisfaction

Despite the eight-figure run rate and explosive growth, there’s no back-patting at Decagon.

Why:

The market is massive. They’ve touched a fraction of a fraction of it. No one feels like they’ve achieved the goal yet.

The final vision: working with the biggest companies out there, deployed across entire surface areas, handling the entirety of customer interactions.

When your goal is that far in the future, it’s easy to stay hungry. There’s always so much more to do.

The Meta-Lesson: Market Selection Trumps Everything

Jesse’s closing advice deserves its own section because it’s the thread running through everything:

Pick the right market.

Within SaaS, if you can find a market with the right properties—clear ROI, measurable outcomes, large TAM, buyer urgency—it makes everyone’s life easier. The structured nature of SaaS (concrete customers, measurable revenue) compounds when you’re in the right market.

His previous consumer company could grow faster but had way less structure, higher variance, and no single customers paying meaningful money that you could talk to deeply.

The lesson: founder-market fit matters, but so does market-outcome fit. Choose a market where the fundamentals work in your favor.

The Q&A: How to Break Into Enterprise Customers

Question from audience: “How do you actually get those first customer conversations? How do you connect with companies like Rippling or Notion?”

Jesse’s Answer:

Find companies known for working with startups. Rippling is one. Then find the shortest path in:

* Maybe your friend knows someone

* Maybe your VC knows someone

* Maybe it’s two hops away

Do it enough and you’ll get through. Maybe the first person isn’t right, but their coworker has a use case. It’s about finding the shortest path and doing it again and again.

When they were starting, they aggressively lined up conversations with literally anyone who would talk to them. That’s just what you have to do.

The Bottom Line

Decagon’s story isn’t about luck. It’s about:

* Choosing a market where ROI is measurable and urgent

* Talking to 100 customers before writing code

* Building a team that reflects founder values and commitment

* Maintaining customer intimacy even as you scale

* Creating an execution culture with clear north stars

* Building for flexibility as AI technology evolves

* Staying hungry no matter how fast you grow

Zero to eight figures in 18 months is extraordinary. But the playbook is surprisingly replicable: pick the right market, talk to customers obsessively, hire for intensity and intelligence, execute relentlessly, and never stop pushing forward.

The AI-native SaaS wave is real. But winners won’t be determined by who has the best models or the most hype. Winners will be determined by who executes this playbook most effectively.

Decagon is showing us how it’s done.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

The State of AI + Software: Where It’s Going - Fast

This deep dive is from Jason Lemkin at the LIVE AI Workshop Wednesday. Sign up here for the next one.

I was talking to a founder recently who’s running at $50 million ARR. Classic SaaS guy turned AI guy. And he’s going to scale from $50M to $100M with just five sales reps and a team of AI agents.

In the old days, at $50 million, you’d probably have at least 100 sales reps. Why? Because to get from $50M to $100M, you need $50M in net new bookings. At $500K net per rep (which is pretty good when you factor in scaling, turnover, and ramp time), you’d need 100 bodies. Minimum.

This founder? Five human reps. Plus AI.

It’s not that he doesn’t need sales reps. It’s not even that he’s not selling. He’s actually doing classic B2B SaaS sales. He’s just doing it with dramatically fewer humans, and for the purposes of this discussion, he’s doing it so much more efficiently. And he has 10,000+ inbound leads a month flowing through this system.

This is where we are right now. And if you’re not paying attention, you’re already behind.

Everything Changed For SaaStr Itself in the Last 180 Days

At the end of Q1 this year, we had zero AI agents in production at SaaStr. Nothing. Nada. We were thinking about it, but we hadn’t deployed anything.

Fast forward to today, and we’ve got:

* Almost 20 AI agents running in production

* Four different AI SDRs deployed and actively working leads

* Salesforce Agent Force rolled out (just started yesterday, so we need a bit more time before sharing all the data)

* An AI BDR from Qualified handling inbound qualification

* A slew of specialized agents for support, research, and operations

It is so much different even on our little team than it was even 100 days ago. And it’s going to keep changing at this pace.

I’ll be honest: we were probably a little behind the curve at the start of the year. Now, we’re kind of at the bleeding edge. And we want to drag everybody along with us because everything’s changing.

The models are changing. The tools are changing. Things that didn’t work last year can work really well now.

Everyone complained about how crummy AI SDRs were last year—and there still are a lot of issues—but now we know how to train them. Now we know how to iterate with them. Now we know how to make them work.

There’s so much more coming, and marketing in some ways is even further behind sales. But it won’t be for long. This whole space is going to radically change in the next 12 months.

More on our AI Agents here.

The Single Most Important Thing You Need to Do to Stay Relevant

Here’s my advice, and I mean this with every fiber of my being:

If you feel behind in AI and go-to-market, the answer is simple: be part of a deployment.

Not just “buy a tool and forget about it.” That teaches you absolutely nothing.

Go buy any tool. Go buy any tool that comes and talks at events. Go buy any leading tool in the space. At some level, it doesn’t even matter which one you start with.

But here’s the critical part: Don’t just buy it. Be part of the deployment.

* Train it yourself

* Be part of the onboarding

* Be part of the errors and the issues

* Be part of the daily iterations you have to do to get it working

Don’t just set and forget. You will learn nothing. You will learn absolutely nothing.

Be part of a deployment to see how it really works. Otherwise, you’ll never really learn.

You don’t have to become a vibe coder like I have. You don’t even need to play with the underlying tech. What you really need to do is get your hands dirty with actual implementation.

The Iconiq Growth Report: 10 Things That Show How Much the World Is Changing

Iconiq Growth—one of the top late-stage B2B growth funds forever, from the early days when they managed Mark Zuckerberg’s family money to being in so many leaders from Canva to Anthropic and beyond—just put out an incredible report.

https://www.saastr.com/iconiqs-state-of-software-in-2025-much-smaller-teams-55-of-gtm-is-now-in-post-sales-2025-is-crushing-2024-for-funding/

It’s almost 100 pages. There’s a lot of detail in there about public companies and nuances that a lot of you don’t really care about. But I picked out what I think are the 10 most relevant things to SaaS folks for B2B that show just how much the world is changing.

A lot of folks are behind here. A lot of folks. And it’s okay—it’s not necessarily fatal. We’re not all being killed by ChatGPT and Anthropic. But the world is changing. This data shows you just how it’s changing and how we all have to adapt.

1. AI Native Companies Are Burning Less (Way Less)

This first data point explains a lot of things that are really confusing on social media. People say, “Oh, these AI native companies, they’re burning all this cash to acquire customers with free tiers and consumption models.”

Actually? Not true.

AI native companies burn substantially less than classic SaaS companies did at the same stage. Why? Because they don’t need to hire armies of people to scale. The unit economics work differently when AI is doing a significant portion of the work.

2. The Sales Team Size Cliff

We’re seeing companies hit $10M, $25M, even $50M ARR with radically smaller sales teams than the old playbook required.

The old rule of thumb: you needed roughly one sales rep per $500K-$1M in ARR, depending on your ACV and sales cycle.

The new reality: AI-native companies are operating with 20-30% of the headcount at equivalent revenue levels.

That $50M company I mentioned at the beginning? That’s not an outlier. That’s the new normal for AI-native companies.

3. CAC Payback Periods Are Compressing

Customer Acquisition Cost (CAC) payback periods are getting shorter in AI-native companies. Not because they’re spending less on acquisition (though some are), but because they’re able to onboard, activate, and expand customers faster with AI-powered customer success and product experiences.

We’re seeing companies hit CAC payback in 6-8 months that would have taken 12-18 months in the classic SaaS playbook.

4. Net Revenue Retention Is Under Pressure Everywhere

Here’s the uncomfortable truth: Net Revenue Retention (NRR) is declining across the board.

Even the best SaaS companies are seeing NRR pressure. Why?

* Competition is fiercer than ever

* Customers are more price-sensitive post-2021

* AI is enabling customers to do more with less, reducing expansion

* New AI-native alternatives are constantly emerging

The days of 130%+ NRR being table stakes? Those are largely over. Now, 120% is excellent. 115% is solid. And anything above 110% in the current environment is worth celebrating.

5. The Growth Multiple Advantage of AI Companies

Here’s where it gets really interesting: AI companies are trading at higher multiples than pure SaaS companies at equivalent growth rates.

Why? Because investors believe:

* The TAM is larger (AI can address bigger markets)

* The margin structure is better long-term

* The defensibility is stronger (data moats, model improvements)

* The growth trajectory is more durable

A SaaS company growing 100% YoY might trade at 10-15x ARR. An AI company growing 100% YoY? We’re seeing 20-30x+ ARR in some cases.

6. Time to $10M ARR Is Accelerating

AI-native companies are hitting $10M ARR significantly faster than classic SaaS companies did.

Classic SaaS: 3-5 years from founding to $10M ARR was typical for fast-growing companies.

AI-native: We’re seeing companies hit $10M ARR in 18-24 months.

Why? Product-led growth combined with AI efficiency. Lower go-to-market costs. Faster time to value. Viral distribution.

7. The Disappearance of the Middle-Market Private Equity Exit

This one is painful but important: The private equity exit path for $20M-$50M ARR SaaS companies has essentially disappeared.

In the 2010s and early 2020s, if you had a solid SaaS company growing 30-50%, your burn rate was low, and your NRR was high, maybe VCs wouldn’t touch you, but a private equity firm would come in and buy you for 6x to sometimes 10x revenue.

You’d get to $20M revenue growing 50-60% and cash flow neutral. No VC was going to fund you there, but a PE firm might buy you for $150M or even $200M.

I don’t see any of those deals anymore. They’ve disappeared.

Why? Because private equity isn’t immune to the fact that:

* Public SaaS companies have seen growth decay

* NRR isn’t as sticky as it used to be

* New AI vendors are putting the old guys at risk

So PE has disappeared from this segment, and VCs have flocked to hyper-growth AI companies.

This may or may not change, but at least don’t live in a dream world where you think the next round is going to come because growth is “pretty good.” It won’t.

8. If You Have Any Doubt About Fundability, Check the Data

We literally added a new benchmarking tool at SaaStr.ai. It’s our third tool there. Go to saastr.ai, click “AI VC” at the top.

Upload your latest investor update or board deck or VC pitch. We will tell you the exact odds you get funded. The exact odds.

And if you don’t like it, don’t shoot the messenger. It’s based on all the recent data from 5,000+ rounds.

I talk to way too many founders who think they’re going to get funded in B2B, and they’re not. So just at least find out. We built the tool. It’s benchmarking. It’s cool.

9. AI Is Creating a New Category of “Efficiency Giants”

We’re seeing the emergence of what I call “efficiency giants”: companies that can reach $100M+ ARR with teams that would have been laughably small in the SaaS 1.0 era.

These companies have:

* Sub-50 person teams at $50M+ ARR

* Gross margins in the 85-90% range

* CAC payback under 6 months

* Growth rates still above 100% YoY

These aren’t just outliers. This is becoming a category.

10. The AI-Native Stack Is Real and Growing

Companies are increasingly building on what I call the “AI-native stack”: infrastructure, tooling, and platforms designed specifically for AI workloads and AI-powered applications.

This includes:

* Vector databases (Pinecone, Weaviate)

* LLM orchestration layers (LangChain, LlamaIndex)

* AI observability and monitoring tools

* Model fine-tuning and deployment platforms

The total spend on AI-native infrastructure is growing 300%+ YoY, creating a massive new category of B2B software.

The Hard Truth About Introducing AI SDRs to Your Team

Someone asked me a great question in the live AMA: “How do you start to thoughtfully introduce AI SDRs to the sales team without scaring off the existing SDRs?”

I was literally dealing with this yesterday. I was at the company all-hands for a company I invested in called MangoMint (SaaS + AI Agents for salons, spas, and doctor’s offices). Great company, great CEO, great culture. People were just so excited at their all-hands.

The CEO asked just one favor of me, “Just do me one favor: don’t scare the SDRs.” He said, “Don’t scare them with all this stuff.”

I said, “Okay, I won’t.”

But actually, I’d already had lunch before the presentation. I sat down at tables with all the SDRs. I’d already had the conversation with them.

They understood that the best have as important a role as ever, and the others don’t.

The VP of sales sat down with me afterward and gave the same message. She said, “We’re aggressively all-in on these tools, and we’re growing. We’re growing triple digits at 8 figures in revenue. This is the future. We have to embrace it.”

So look, here’s the hard truth: If your SDRs are going to quit because of AI, they’re going to quit anyway.

And the tenure of an SDR is very short to begin with. Just be clear: we need everybody who can perform.

It was funny. We were sitting there at lunch with the SDRs, and one of them was one of the best ones on Marshall’s team. I’m sitting next to her, and she’s already got her laptop up. She’s scanning a salon that she’s doing outbound to during lunch. She’s literally doing true outbound research during lunch.

She’s got a job forever.

The SDRs that just want to run random emails and never learn the product? They don’t.

I think we’re past the point where you have to worry about scaring the team. Be kind. Make sure everyone who crushes it knows that everyone who is A-tier on your team has a role. That hasn’t changed.

It’s even harder to find A-tier folks now. But if folks don’t want to adapt, they won’t have a future. And I think you’re better off being straightforward with them.

Just do it.

Here’s the other thing: As soon as you actually roll out the tool, someone’s going to quit anyway.

This is the story with every company I’ve invested in. Even at SaaStr, on our little team, we rolled out a tool called Momentum (or it might have been from Attention—they’re great too).

Even on our little team, the day we rolled it out, someone on our sales team quit. The day we rolled it out.

Why? Because now all his actions, all his data were going to show up in real-time in a report and in Salesforce. You couldn’t hide. You couldn’t pretend you did five calls this day when you didn’t. You couldn’t pretend you did the work when you were out at this tool.

Somebody quits. And honestly? It’s all for the best. The truth is, it’s all for the best.

We’ve got to get out of kumbaya land because this is a new high-growth world. You’ve got to be a part of it.

The Real Cost of Implementing AI: What You Actually Need to Budget

Another great question: “How much should we be estimating for cost to invest per year in AI to get started?”

This is better than it sounds, and the answer is going to surprise you.

Let me give you the SaaStr example. We’re tiny. We spend maybe $10,000 a year on Salesforce. Maybe a little more. And the same or a bit more on Marketo, which we essentially use as a CRM for our community (Salesforce is our CRM for sponsors and ‘sales’).

I’ve been a Salesforce customer for 20 years. First piece of SaaS I ever purchased for real. I never paid for anything before Salesforce.

But here’s the thing: we spent maybe $10K on Salesforce notionally.

We spent $500,000 on our AI agents across these 21 agents. $500,000.

Think about that ratio for a minute. There are so many learnings in that ratio.

It’s why Salesforce has to own Agent Force. They have to win. It’s why HubSpot has to win here.

But we are spending far more on agents than we are on CRM. That’s a meta thing to think about.

The Real Cost Structure of AI Tools

The second thing to think about: these apps that need to be trained with forward-deployed engineers and hands-on work? I don’t think very many of them are less than $30,000 or $50,000 a year.

A lot of them actually try to have a price point that’s approaching $100K—like $60K a year base, and then $20K or $30K for the forward-deployed engineer and training.

Sometimes they include it, sometimes they charge for it. Whether they do or don’t, it ain’t free to do it right.

For me, I would certainly subsidize it if I were a vendor, but it costs that amount. Having a really great technical resource help you train your AI for a couple of weeks? That’s not free. It’s pretty damn expensive.

And this relates to the meta issue: You’ve got to invest the time.

There’s a reason these apps cost $50K to $100K. They are a lot of money. Be wary of super cheap apps. Be wary.

It’s not that we’re not getting there. It’s not that everything isn’t going to get better. But you can’t cut corners on training.

If instead of $50K a year, you’re buying a solution that’s $500 a month or $50 a month, I’m not saying it’s not going to work, but you’re going to have to do even more training. You better take on a massive amount of burden to train this app because you’re not getting the benefit of the forward-deployed engineers and all the training, tuning, warming, and everything the vendor does.

There’s no shortcut here. There is no shortcut.

It is an existential issue. This will change over time, but right now, the agents can’t train the agents themselves the way the best humans can.

It’s tough to get away with less than $30K-$50K for any of these tools when a month of training is required—or at least several weeks.

What This All Means for You

Look, I get it. This is overwhelming. The pace of change is brutal. Every time you think you’ve figured something out, the models get better, the tools evolve, and the playbook shifts.

But here’s what you need to take away from all of this:

1. The efficiency advantage is real. AI-native companies are operating with dramatically lower costs and higher margins than classic SaaS ever could. If you’re not thinking about how to capture this advantage, you’re leaving money—and competitive positioning—on the table.

2. You can’t learn by observing. You have to deploy. You have to train. You have to iterate. Reading blog posts and attending webinars isn’t going to cut it. Get in the arena.

3. The funding environment has bifurcated. Hyper-growth AI companies are getting funded at eye-watering valuations. Solid but unspectacular SaaS companies are getting passed over. Know which category you’re in and plan accordingly.

4. Your team structure needs to change. The old GTM playbooks are dying. You don’t need 100 SDRs to scale to $100M anymore. But you do need the right people who can work alongside AI, train it, and amplify their effectiveness with it.

5. Budget for this properly. You’re going to spend more on AI agents than you think. And that’s okay—the ROI is there. But don’t go into this thinking you’re going to deploy enterprise-grade AI solutions for $500/month. You’re not.

6. Speed matters more than ever. The companies that figure this out in the next 12 months are going to have a massive, compounding advantage. The companies that wait are going to find themselves 18-24 months behind, and in this market, that might be fatal.

The Bottom Line

We’re in the early innings of the biggest shift in B2B software since the move from on-premise to cloud. Maybe bigger.

The companies that win are going to be the ones that:

* Deploy aggressively

* Learn constantly

* Adapt their teams and processes

* Invest in the right tools and training

* Move fast without breaking (too many) things

The companies that lose are going to be the ones that watch from the sidelines, wait for “best practices” to emerge, and try to protect the status quo.

Don’t be the second group.

Get in. Deploy something. Train it. Break it. Fix it. Learn from it.

That’s the only way forward.

And if you want to learn more about how we’re doing all of this in practice, come to SaaStr AI in London on December 1-2. We’ll have speakers from OpenAI, Whiz, Clay, Intercom, Fin, and all your favorite B2B + AI companies.

We’ll do live sessions. We’ll share real data. We’ll show you what’s working and what’s not.

Because this is changing every quarter, and we’re all learning together.

Let’s go.

Want to see if your company is fundable in today’s market? Check out our AI VC benchmarking tool at saastr.ai. Upload your deck and get your exact odds. It’s free. It’s data-driven. And it might save you months of wasted pitching.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

A deep dive into the playbook, lessons learned, and brutal truths about deploying 20+ AI agents into production — that actually work.

During Dreamforce Jason (CEO) and Amelia (Chief AI Officer) at SaaStr dropped by Qualified’s office for a deep dive on its AI Agents and top learnings with Kraig Swensrud (Founder & CEO, Qualified)

The Bottom Line Up Front

If you’re a CMO, CRO, or founder and you haven’t deployed at least one AI agent by Halloween 2025, you’re already behind. Not “might fall behind” — you’re already behind. “Every VC I know where a startup hasn’t made the jump yet has given up hope on that company. That’s not hyperbole. That’s the market reality.” per Jason.

But here’s what under-discussed training is more important than picking the perfect vendor. We started 2025 with zero AI agents at SaaStr. Now we have 20 in production. The secret wasn’t finding magical tools — it was investing 30 days of deep training upfront, then maintaining an hour every single day.

Here’s exactly how we did it, what worked, what failed, and the framework you need to deploy agents that actually drive revenue.

The New Budget Reality: Why Traditional SaaS Playbooks Are Dead

Let’s start with the uncomfortable truth about 2025 budgets:

Traditional SaaS budgets are frozen. CEOs are going around the table telling every functional head to cut 20-30% of their SaaS apps. Half of incremental budgets are going to price increases — Salesforce raised prices 8% this year, others 6-7%. When your IT budget is growing 6% but your core vendors are raising prices 7-8%, where’s the room for another business process workflow app?

But AI budgets are exploding. Business software is growing faster than it has ever before — if you tap into AI budget. It’s the only incremental budget most companies have. Nobody is putting more money into old SaaS software. It’s all going into AI.

This creates a tale of two cities:

* Classic SaaS is geriatric ... but ...

* B2B software powered by AI is exploding

If you’re selling the way you sold in 2021, with the 2019 Marketo playbook, there’s no budget for you. The playbook doesn’t work. But tell that to anyone on fire with AI — everything works. Outbound works. Events work. Meeting with customers works. If anyone wants to buy your product, it all works.

The Vendor Selection Myth: Training Trumps Everything

Here’s the question Jason gets constantly on LinkedIn: “What’s better — Replit or Lovable? Which AI SDR platform is best? What AI tool is best for RevOps?”

Wrong question.

Here’s what Jason learned after deploying 20 agents: Pick a leader and go deep. Training is more important than vendor selection.

There are very few agentic products with any level of complexity where you can just flip a switch. Take Qualified — it set up seven appointments for us on its own. Sounds great, right? Put it in a case study. But it wouldn’t have happened if we put zero minutes of training into it.

The New Deployment Reality

Remember how people bought enterprise software before AI? You’d buy it from a sales rep. You might hire Accenture or Blue Wolf to deploy it over a year. You’d hope and pray it worked like the sales rep told you.

No one will tolerate that in AI this year. Time to value often has to happen before you even get a contract signed.

Jason is 100 days into Replit but probably 200 hours expert now to make it great. That’s a sea change in how we use software. Companies are used to buying from a people person and scaling up over time with humans doing all the work. Those days are behind us.

The Training Investment Framework

For any agent you deploy:

* Commit 30 days upfront — train it every single day

* Then commit weekly — every week after that

* Budget 1 hour daily ongoing — this is your new normal

Most people don’t know how to do this. But if you don’t train it, you won’t get ROI. Period.

The SaaStr Agent Journey: From 0 to 20 in 10 Months

Agent #1: Jason AI (The Horizontal Foundation) aka “Digital Jason”

SaaStr started with Delphi, building a clone of Jason — “Jason AI.“ It ingested all of SaaStr content (20 million words, 1,000 YouTube videos, all tweets, all LinkedIn posts). It actually broke the ingestion engine at first, but once it worked, it was magical.

Key insight: SaaStr started with a horizontal agent that could do a little bit of everything. No complex workflows. Just digital Jason answering founder questions. Training was broad, but not especially deep. Perfect for a general agent, answering general questions.

What happened: People started using it for things we didn’t envision. They used it for support questions about attending our events, sponsor questions, booth logistics. Classic issue with AI agents — if it’s good, people will start using it for everything. If they trust it, usage expands.

Then it started to hallucinate on event-specific questions. So the team uploaded the prospectus. With that additional context and training, it got good. Maybe 20% as good as Qualified, but 20% is infinitely better than zero.

Before: Pre-Fin Intercom with week-long response timesAfter: Instant answers, 90% accuracy on customer questions

The Stair-Step Strategy That Actually Worked

Here’s the deployment sequence that worked for SaaStr:

Step 1: Horizontal Agent (Delphi/Jason AI)

* Easiest to deploy

* Least training required

* Built confidence

* Got an early win

Step 2: Outbound SDR (Artisan)

* Lower risk than inbound

* SaaStr’s SDRs wouldn’t do outbound anyway

* Better to send a better email via agent than mediocre human

* Less risk than trusting agents with precious inbound leads

Step 3: Inbound Qualification (Qualified)

* Higher complexity

* More critical to get right

* But SaaStr’s qualification process was nearly broken

* Low bar made deployment easier

Step 4-20: Specialized Agents

* Email campaigns

* Website traffic deanonymization

* Follow-up sequences

* Support ticketing

* AI speaker content review

* AI VC

Why Stair-Stepping Matters

If you fail with your first agentic thing, you won’t know what went wrong. You’ll get frustrated. You won’t build the confidence to keep going.

SaaStr accidentally stair-stepped their deployment, and it’s why they succeeded. They gained confidence with each win. By the time they deployed Qualified for inbound (higher stakes), the team knew how to train agents properly.

The framework:

* Use horizontal agents where bar is low, where there is nothing already and you have some strong content to train it on. But it doesn’t need to be a vast amount of content. It just has to be useful and relevant.

* Use specialized agents at first for areas with zero human coverage. Because the bar is low for success.

* Save the hardest, most critical deployments for when you have 2-3 agents under your belt.

The “Layup Roles”: Where to Start Right Now

The biggest question: where do I actually start?

The layup roles are where there’s a trustworthy tool that exists, but work is just not getting done in your org.

There are lots of reasons work doesn’t get done:

* Tough to hire in that area

* Tough geography for hiring

* Not your DNA (product-centric companies suck at sales, sales-centric companies suck at product)

* High turnover

* Junior team members who don’t know your product/market/competitors

Don’t try to replace stuff that’s working. Look at where there’s literally no one doing the work.

The Proven Starting Points

#1: Customer Support (The Easiest Layup)

Go try your own support as a founder or executive right now. Nine times out of ten, you’ll cry at how bad it is. Submit a P1 issue. See if you get “We’ll get back to you in a week” or “I don’t understand the problem.”

That’s where to start. The bar is so low, your agent doesn’t have to be amazing. It just has to be better than terrible.

We saw this at SaaStr: Went from Intercom (week response times) to instant accurate answers in 90% of cases. The other 10%? The agent knew to escalate.

#2: Outbound SDR/BDR

For us, this was lower hanging fruit than inbound. Why? Our SDRs couldn’t (or wouldn’t) do outbound consistently. The bar: send better emails than mediocre humans were sending — or in our case, not sending.

Less risky than inbound because you’re not trusting agents with precious inbound leads. You’re just trying to generate pipeline that wasn’t being generated before.

#3: Inbound Qualification (Higher Complexity, But Massive ROI)

SaaStr’s process before Qualified: come to website, fill out contact form, hope over the next week a human gets back to you. Maybe they follow up, maybe they don’t. Round-robin meant inconsistent quality.

The Qualified deployment results:

* 7 appointments set autonomously in the first weeks

* ~100 tickets sold to London event in 6 weeks

* Overnight coverage (Pacific team sleeping while UK prospects active)

* Better qualification than our human process

The bar wasn’t “be better than a great team of 10 product experts with 5-minute SLAs.” The bar was “be better than broken.” That made deployment possible.

What About the “Hero Purchase”?

Here’s the failure mode I see CMOs make constantly: they want a hero purchase. They want to go to the CEO and say “I bought AI.”

This will probably fail. Why?

* That hero purchase is likely aligned with something you’re already decent at

* The bar is high because it’s already working

* It’s your first agent, so you don’t know how to train it

* You’re trying to make something 20% better instead of fixing something that’s 0% done

Better approach: Leave that hero app on the shelf. Deploy something where work literally isn’t happening. Get a win. Build confidence. Then deploy the hero app when you know what you’re doing.

If AI is only going to make something 20% better, maybe 2025 isn’t the year. Maybe that’s a 2026 project.

The Daily Reality: What Managing 20 Agents Actually Looks Like

The hour Amelia spends with agents every morning used to be spent with people.

Before: One-on-ones with SDRs and BDRs, going through accounts, providing context on leads, giving them history on prospects.

Now: Amelia wakes up, checks each agent:

* Qualified — see meetings booked overnight (London prospects while we sleep), review conversations, give context to sales team

* Artisan — review outbound performance, adjust sequences

* Delphi — check Digital Jason AI conversations, refine responses

* Agent Force — review support tickets, catch edge cases

* Cycle through remaining agents

Total time: 1 hour

What’s different:

* Higher cognitive load — you have to think more, use more brain cells

* Compounds over time — unlike humans who leave, every hour of training pays forward

* Agents don’t cry — they don’t get emotional, but they require deep thinking

* You can do more — now expanding to email campaigns, deanonymization, follow-up sequences

The Early Days Reality

First 30 days: Way more than 1 hour. You’re setting everything up, fixing data issues (our Salesforce data was terrible — agents expose that), connecting systems, training workflows.

After 30 days: Maintenance mode at ~1 hour daily, but you’re constantly expanding use cases.

The Compound Effect

With humans: Train for months, they get good, they leave, repeat.

With agents: Every hour of training compounds. They get better every day. They don’t forget. They don’t quit. They work 24/7.

Example: Qualified handles UK prospects overnight while Pacific team sleeps. Those conversations used to just... not happen. Now they do, and they’re high quality.

The Brutal Truths About Team Reactions

For Founders: Some People Will Leave When You Roll Out AI Agents. That’s Okay

Amelia was talking to a founder this week at Dreamforce. She has some agents deployed but is worried about rolling out to her 50-person sales team. “I think they might revolt. Some might quit.”

Amelia’s answer: That’s okay. Let them.

It’s almost 2026. If your team can’t get over the hurdle, some may quietly leave. You cannot forsake the future of your company to avoid hurting feelings.

Rip the band-aid off. Do it the right way — show them they can be empowered with AI, crush quota with AI, do more with AI. The best ones will see this and get excited. They should be roaring to get AI going.

If you see people hiding in the corner, maybe it’s okay if they quietly weed themselves out.

The Exposure Problem

SaaStr had someone leave their team the day they hooked up Momentum (their conversation recorder and GTM tool). Why? It Slacks summaries after every sales call and notifies Amelia when opportunities open or meetings book.

This rep had zero activity. The system exposed it instantly. They quit that day. You can’t hide anymore.

For Individual Contributors: Deploy It Yourself or Find a New Job

Jason recently had a CRO he’s worked with for years reach out in a panicked email: “I need to learn AI. I’ll do anything. I’ll intern for you. I’ll hang out in Amelia’s office.”

Jason’s advice: Don’t learn it. DO it.

Buying a subscription to Claude does not count as learning AI. Here’s what counts:

* Pick one agentic product — simplest possible use case

* Buy it yourself — don’t delegate to your team

* Set it up yourself — okay to get help with CRM connectors, but own the deployment

* Train it yourself — take your time, do something simple

* QA and test yourself — hands on keyboard

If you deploy an agent yourself, train it, QA it, test it — you will be ahead of 90% of the world.

The CMO Who Jason Couldn’t Help

Another CMO Jason loves reached out after many years. Jason had gotten her the last two jobs. She was ready for number three.

Jason’s response: “I got nothing for you.”

Why? She doesn’t know this stuff yet. She hasn’t deployed an agent. She can’t tell Jason how training went, what the issues were, what worked.

Here’s the weird thing: There is much more demand for great CMOs than CMOs think. Jason knows 100 CEOs today who are desperate for a great marketer. They’ll do anything.

But they do not want someone running the 2019 Marketo playbook. Zero interest in hiring that person.

Jason’s advice to her: “Go deploy an agent. Tell me how it worked. Tell me how training went. Tell me the issues. Come back to me. I’ll get you two jobs.”

The ROI Reality: What Results Actually Look Like

Qualified Results (6-8 Weeks of Deployment)

* 7 autonomous meeting bookings per week in early weeks

* ~100 tickets sold to London event so far out of ~2,000 total (5% conversion via agent)

* 24/7 coverage — overnight conversations with UK prospects while Pacific team sleeps

* Better qualification than human process — knows when to escalate, provides context to sales team

Artisan Results (Outbound)

* Better email performance than human SDRs who weren’t doing outbound

* Consistent execution — no more “round robin to rep who doesn’t follow up”

* Scaling sequences we could never scale with humans

Overall Impact

* From 0 to 20 agents in 10 months

* ~12 core agents in production running go-to-market

* 1 hour daily maintenance (down from way more in early days)

* Work that literally wasn’t getting done now happening 24/7

The brutal truth: Half the time, SaaStr’s sales team wasn’t following up with inbound leads. At least half the time. With round-robin, good reps followed up instantly. Mediocre reps... maybe, maybe not.

Now: Instant response, 24/7, consistent quality, full context provided to sales team.

The Deadline Reality: How Fast Do You Need to Move?

For Startups: Now

If you’re a startup and you don’t have anything disruptive in production by now, you’re already behind.

Every VC Jason knows, every investor he knows — if the startup hasn’t made the jump yet, they’ve lost hope. Not “losing hope” — they’ve LOST hope.

Why?

You could make excuses when ChatGPT came out in 2023. There were so many hallucinations. We could all make fun of it.

2024? You could be a critic. Maybe you’d start losing deals to hot AI companies, but at least you could have a principled position that this stuff wasn’t going to work.

2025? There’s no excuse.

Everything’s great now. Even mediocre software got 5x better when you put Claude 4 in. Replit took 10 years to get to $1M ARR. They added Claude 4, now they’re almost $200M in one year. Same IDE. Same software.

If you’re not in market with a disruptive AI agent by the end of the quarter:

You need a brand new team. Maybe 80% have to go. Give them a nice Thanksgiving bonus and a turkey and tell them they’ve got to go. Because you’ve had 10+ months.

For Enterprises: It Depends (But Don’t Wait Long)

If your business is fine, if growth is as good or better than 12 months ago, if you’re in a traditional industry where everything’s working — you can be a late adopter.

There are late adopter advantages in AI:

* Tools get better fast (Replit is epically better than 100 days ago)

* Training resources improve

* Best practices emerge

* Edge cases get solved

But if you’re facing external pressure, if competition is heating up, if you’re losing deals — you need to move now.

The New Software Buying Reality

In the old SaaS days, a customer would come in and say “I need this feature.” You’d promise it in the contract even though it didn’t exist. You’d brute force it over 12 months.

Those days are dying in AI.

Don’t promise me it’ll be here in a year. Show me the agent works today. If it’s not ready, I don’t blame you. I get it’s early. Come back to me when it works. It might be ready in three months.

Time to value has to be before contract signature in many cases now.

The Framework: Your 90-Day Deployment Plan

Days 1-30: Foundation and First Agent

Week 1: Pick Your Layup

* Audit what’s NOT getting done (support? outbound? qualification?)

* Pick the simplest use case with lowest bar

* Select a leading vendor (doesn’t have to be THE leader, just A leader)

Week 2-3: Setup and Integration

* Buy the tool yourself (don’t delegate if you’re the exec)

* Set up core integrations (CRM, email, support systems)

* Clean your data (this will take longer than you think)

* Build basic workflows

Week 4: Daily Training

* Train the agent every single day

* Test edge cases

* Refine responses

* Build your QA process

Success metric: One working agent handling real work by day 30

Days 31-60: Expansion and Specialization

Week 5-6: Monitor and Optimize

* Spend 1 hour daily with your agent

* Review conversations/outputs

* Provide additional context

* Expand use cases slightly

Week 7-8: Agent #2

* Pick second use case (slightly harder)

* Apply lessons from Agent #1

* Faster setup because you know what you’re doing

* Stair-step to more complexity

Success metric: Two agents in production, 1 hour daily maintenance

Days 61-90: Scale and Specialize

Week 9-10: Vertical Agents

* Move from horizontal to specialized

* Deploy agents for specific workflows

* Start tackling harder use cases

* Build your agent management system

Week 11-12: Team Enablement

* Train your team on working with agents

* Build handoff processes

* Create monitoring systems

* Document what works

Success metric: 3-5 agents in production, team enabled, clear ROI

The Future: What’s Coming Next

Voice and Video Agents (Live at SaaStr Now)

SaaStr is deploying Amelia AI with voice and video. Not just text chat — full video agent of Amelia on their event sites.

Why? Trust has been built through chat. Now the team can do more:

* Overnight support with full video presence

* Event co-pilot for attendees

* Session recommendations

* Networking coordination

* “Real Amelia is asleep, but Amelia AI is awake”

The funky part: People walk up to real Amelia now and say “You’re that Amelia AI!” The line between digital and real is blurring.

The Compounding Use Cases

Once you have 5-10 agents working well, new use cases emerge constantly:

* Email deanonymization of website traffic

* Follow-up campaigns for unconverted prospects

* Pre-meeting research and briefings

* Post-meeting follow-up and deal progression

* Internal knowledge management

* Onboarding automation

The cognitive load increases, but so does output. It’s exhausting in a different way than managing people, but it’s more productive.

The Market Reality: Everyone’s In Market

Traditionally, you’d hope a prospect would be in market every 5-6 years for a big buy. Why? Takes a year to get going, second year to scale, third year to get frustrated, fourth/fifth year to start looking at new solutions.

Right now, everything’s in market again.

This hasn’t happened since maybe when the web started. Everyone either has AI budget or is under pressure from their board/CEO to show AI innovation. It’s the only incremental budget.

For vendors: This is the opportunity of a lifetimeFor buyers: The tools will never be better-priced or more hungry for your business

The Final Word: Just Rip the Band-Aid Off

Let’s be brutally direct:

If you’re a go-to-market executive and you haven’t deployed an AI agent yourself by now, you’re unemployable.

Not “will be unemployable” — you ARE unemployable. Jason has CMO friends he loves, who he’s gotten jobs for multiple times, who he now has nothing for. Why? They don’t know this stuff yet.

The Simple Advice:

* Don’t learn AI — DO AI

* Pick one agent, simplest possible use case

* Deploy it yourself (hands on keyboard)

* Train it for 30 days

* Spend 1 hour daily after that

You will be ahead of 90% of the world.

The agents don’t cry. They work 24/7. Every hour of training compounds. They get better every day, they don’t forget, they don’t quit.

But only if you actually do it.

It’s almost 2026. Rip the band-aid off. Let’s go.

Key Takeaways

* Training > Vendor Selection — Pick a leader and go deep, training matters more than choosing the perfect tool

* Stair-Step Your Deployment — Start simple, build confidence, then tackle harder use cases

* Find the Layup Roles — Deploy where work isn’t getting done, not where it’s already working

* Budget 30 Days + 1 Hour Daily — This is your new reality, embrace it

* Halloween Deadline for Startups — If you’re not in production now, you need a new team

* Do It, Don’t Learn It — Hands on keyboard deployment is the only way to actually learn

* Some People Will Leave — That’s okay, you can’t forsake your company’s future

* The Compound Effect is Real — Unlike humans, agents get better every day forever

The future belongs to companies that move now. Not next quarter. Now.

Want to see Jason AI in action? Go to SaaStr.com and click the bottom right corner. Want to see how SaaStr deployed 20 agents? Follow Jason Lemkin on LinkedIn or come with your questions LIVE to SaaStr AI London on Dec 1-2!!

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com

I think after 100+ days and with 9 apps vibe coded into production with @replit used over 500,000 times we’re just getting going. And … I think key to that is that I’m now “Replit Fluent”.

What does that mean? It’s a state where I know how to vibe well enough (without a developer), and I know the app and its capability and limits well enough, that I can basically see any app I want to build in my head, and now know how to completely prompt it and shepherd it to production … before I start.

I can now will almost any ‘normal’ app into existence that I want to build. Without a developer.

I remember in the early days of Cursor my son actually paid out of his own pocket for the first time since ChatGPT (he’s awfully smart). He said back then Cursor could now do 90% of his coding for him, but “for most people, it might be 10% or less.” I didn’t get what he meant at the time, but today I do. It’s more than being “good at prompting”. It’s understanding the system well enough to already understand its outputs, its limitations, and exactly how it works in practice — before you start. It’s being truly fluent in the agent.

This isn’t to say I don’t still have bumps and that some things end up harder or longer than anticipated. But pretty much now I can will most things into production on Replit very predictably with the agent — because I can already plan them out fully in my mind before I first “prompt” the agent.

What that means in practice is 3 things:

First, I 100% know if a project will work now before I start in Replit. If I can see it to completion in my mind, now I can finish it to a reasonably high standard. That is a huge boon. When I started ‘prosumer’ vibe coding all of 100+ days ago :), I couldn’t finish my first project. In part, it was because I picked a very complex product to start. But there are many stories of others in similar boats. They can’t finish their vibe coded app. But now — I have a 100% chance of finishing a project, and in roughly the amount of time I budget for it in mind before starting.

Second, now I’m merely just time constrained in what I can build. I already have a couple of jobs. I have $100m+ to invest in fresh capital at SaaStr Fund, and running SaaStr itself is an eight figure business. Both take a lot of time. But I set aside about 1.5-2 hours a day to vibe code. That’s my budget. That’s what I can build now.

Third, maintenance and new features and upgrades to existing apps I’ve vibe’d consume more and more of my time budget. This of course is true of any software. It just catches up to you with prosumer vibe coding. So now that I can basically truly build anything I want, the question becomes — do I have enough time to make it great? Or should that time go into making my existing 9 apps even better? I now can make a pretty good v1 of anything I want to. But getting to great takes time. It always has.

I think it’s a big deal to be able to get to this state without a developer and without being a developer. Yes, I co-founded a B2B startup that went from $0 to $200m+ ARR so I had that going for me (which is a lot). But the fact I am now “Replit Fluent” and that now my only limit on building is my own time … is super interesting.

The world is so, so much different than 12 months ago.

There is a learning curve in AI. In almost every great tool and every great agent, at least for now. Don’t let anyone tell you there isn’t. Ignore any marketing that tells you there isn’t. If you want to get most AI Agents into production, e.g. a great AI SDR or great AI Support, and have it work well — you have to learn and often train the AI. IMHO you also have to learn tools like Replit to truly get out of them what they promise. You will be much, much better 100 hours in.

But if you get to a State of Fluency in a prosumer vibe coding app … it’s like a superpower you never had before.

Thanks for reading SaaStr AI: How To Sell, Scale, and Win! Subscribe for free to receive new posts and support my work.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cloud.substack.com