Listen now on YouTube, Spotify, and Apple.

—

Joe Wilkinson helped scale Lucid past $100M ARR and now advises early-stage startups on product-led growth and turning visitors into customers through his boutique consultancy Artisan Strategies.

Here’s some of my takeaways from this week’s episode…

1/ 🎯 Value First, Distribution Second• The best products create an amazing core experience that sells itself• Distribution becomes about showcasing that incredible experience• Start with making something people genuinely love, then figure out how to reach them2/ 🔄 Monthly > Annual Early On• Monthly subscriptions force everyone to focus on customer retention• Teams naturally optimize for keeping users happy and engaged• Build better products by allowing customers to "vote with their feet"3/ 📊 Data Should Serve Action, Not Vice Versa• Data alone never makes decisions - humans do• Avoid using data as an excuse for inaction• Better to ship without perfect data than to have perfect data and never ship4/ 🎨 Not All Clicks Are Equal• Remove friction that blocks value, not just any friction• Video games have long onboarding but users don't mind because it's valuable• Focus on maintaining momentum toward the core value proposition5/ 💼 The Power of Fractional Leadership• Get higher-caliber talent than you could afford full-time• Perfect for early-stage companies that need expertise but not full-time roles• Can be more impactful in 1 day/week than a full-time junior hire6/ 🎯 Operational Risk > Opportunity Risk• Focus on executable strategies with clear value vs speculative bets• Example: Master proven channels (SEO) vs chasing new unproven ones• Success comes from execution quality, not gambling on opportunities7/ 🌱 Solve The Now, Build For Later• Must survive today to thrive tomorrow• Make decisions that work now but inform future vision• Balance immediate revenue needs with long-term product development8/ 🎮 Product-Led Growth Requires Consumer-Grade Experience• Enterprise software needs to feel as good as consumer apps• Focus on individual user delight first• Let grassroots adoption drive enterprise sales

—

Where to find Joe Wilkinson:

* LinkedIn: https://www.linkedin.com/in/josephwilkinson/

* Web: https://artisangrowthstrategies.com/

—

In this episode, we cover:

00:00 Global Connections: The Power of the Internet 00:25 The Evolution of Podcasting and AI 01:55 Insights from Venture Capital: Learning from the Best 06:35 Growth Strategies: Lessons from Lucid 11:19 The Journey of Lucid: From Startup to Success 13:11 Lucid's Core Offerings and Innovations 13:58 Mindset for Growth and Experimentation 16:10 The Importance of Testing and Learning 17:07 A/B Testing Success Rates and Strategies 19:47 Achieving Statistical Significance in Testing 23:44 Frequentist vs Bayesian Approaches to Testing 30:03 Understanding Click Value in User Onboarding 34:17 The Concept of Product-Led Growth 36:02 Grassroots vs Traditional Sales Models 40:09 Common Misconceptions about Product-Led Growth 41:34 The Role of Data in Business Decisions 42:30 The Role of Data in Decision Making 47:21 Understanding Time Horizons in Business 50:24 Navigating the Middling Problem 57:11 The Rise of Fractional Work 01:03:51 Operational Risk vs Opportunity Risk

—

Obligatory disclaimer: I've worked at YouTube and Google for about a decade in various marketing teams. Nothing I say in my personal spaces is necessarily endorsed by them.

Listen now on YouTube, Spotify, and Apple.

—

From studying mechanical engineering to leading Product at startups that have raised a combined $500 million, Kari Ostevik has deep experience shipping successful products in high-stakes environments. And from her time leading growth after a CMO left, Kari has a unique perspective on the Product/Marketing relationship.

Some takeaways:

1. 📢 Push Past the Squeaky Wheel Trap: In many orgs, decisions often default to "whoever shouts loudest." Combat this by bringing hard data + user interviews to the table.

Pro tip: Build a compelling narrative around why certain features deserve priority over what your noisiest stakeholder might want.

2. ⏱️ Meeting Hack Alert: Nobody reads pre-reads (shocking, we know). Instead of sending docs ahead, block dedicated "heads down" time during meetings.

Kari's move? Put on lo-fi beats and give everyone 5-10 mins to process + comment. Way more effective than hoping people did the homework.

3. 🧠 The "Crazy Eights" Brainstorming Tech: Need fresh product ideas? Fold paper into 8 squares, give everyone 8 minutes (1 min per square) to sketch solutions.

The magic? People work independently first = less groupthink, more innovation.

Bonus: Works for remote teams via Miro.

4. 🎯 Stop Being Data-Obsessed (Or At Least Pick The Right Metrics): While metrics matter, Kari warns against getting too tunnel-visioned on short-term data wins.

Hot take: Sometimes chasing metrics can actually wreck your user experience. Balance quantitative insights with maintaining product vision.Like we saw earlier with Sundar, consider defining better metrics that are more aligned with your long-term vision.

5. 👑 The Authority Paradox: PMs lead without direct authority (awkward). Kari's solution? Early 1:1s with team members to understand their past PM experiences + explicitly position yourself as their "umbrella" – there to help them do their best work.

6. 🎥 The Gong-Guerrilla Testing Matrix: Use this to identify what really matters to your decision making users (aka, ideal customer profile / ICP).

* B2B Move: Mine Gong (sales call recording) data for deep user insights

* B2C Move: Hit the streets for rapid guerrilla testing with random folks

Key difference? B2B requires more finesse (sales team gets nervous about customer access), while B2C lets you move fast and break things.

7. 📐 The "Two-by-Two" Priority Framework: Skip complex prioritization matrices. Plot features on two axes: ease vs. value. The real gold? It forces explicit conversations about why something ranks higher/lower than other options.

—

Where to find Kari Ostevik:

* LinkedIn: https://www.linkedin.com/in/kariostevik/

* X: https://x.com/kariostevik

—

In this episode, we cover:

00:00 Welcome 01:05 Transitioning Through Major Company Changes 02:40 Insights from B2B vs B2C Product Management 06:03 Prioritization Techniques in Product Development 09:36 Effective Cross-Functional Collaboration 11:34 Innovative Workshop Techniques for Product Teams 14:13 Collaborative Ideation Techniques 15:06 The Importance of Data in Product Management 15:21 Balancing Data-Driven Decisions with User Experience 16:58 Staying Connected to Industry Best Practices 20:49 Managing Teams and Building Relationships 21:19 Lessons from Teaching and Leadership 24:19 Advice for Aspiring Product Managers 24:33 Lightning Round Insights

—

Obligatory disclaimer: I've worked at YouTube and Google for about a decade in various marketing teams. Nothing I say in my personal spaces is necessarily endorsed by them.

Listen now on YouTube, Spotify, and Apple.

—

Adam Miller has held key leadership roles at Uber, Postmates, Turo, and beyond, often with a title that looks something like “director of user acquisition”. Adam is known for his data-driven approach to marketing, his expertise in optimizing customer acquisition costs, and his ability to navigate the complexities of multi-sided marketplaces. Currently, he advises and invests in early-stage marketplace companies.

Some takeaways:

1. 💸 The Zero-Interest Hangover: LTV (Lifetime Value) calculations were a product of the zero-interest rate era. Today's reality? Focus on shorter payback periods and sustainable unit economics. The days of "10-year customer lifetime" projections are over, folks.

2. 🔋 Supply Side = Life Blood: While everyone obsesses over demand, Miller says supply management is the real MVP. Without it, you're just "a bunch of people looking for something they can't possibly buy." Start there, then worry about customers.

3. ⚖️ The Utilization Sweet Spot: Perfect marketplace utilization isn't 100% — it depends on your model. Commoditized services (like Uber) can push higher, while unique inventory platforms (like Airbnb) might thrive at 15-20%. The key? Match your utilization targets to your market type.

4. 🚀 Cold Start Strategy: Don't build supply from scratch. Find existing supply that's underutilized (think: black car services pre-Uber) and make it a no-brainer for them to join your platform. Zero friction + zero cost = early adoption.

5. 🤖 AI's Marketplace Impact: While AI will disrupt many advantages, it won't touch network effects. Translation: Build a strong network and you'll have an "unnatural advantage" in the AI era. Sorry, AI won't help someone build the next Uber overnight (though it might be able to code it).

6. 🎬 The New Video Playbook: Social video is getting shorter, but Miller warns against oversimplifying complex stories. His advice? Take the time needed to "make something big, simple" rather than forcing everything into a 15-second slot.

7. 🎯 Modern Marketplace Musts: Three non-negotiables:

* Supply management (onboarding + retention)

* Trust and safety (the foundation of everything)

* Liquidity (right supply + right demand + right place = 🎯)

8. 📈 The V-Shaped Marketer: Forget being T-shaped. Start deep in one area, then expand into a V-shape by building multiple deep expertise areas over time. Call it the "sequential excellence" approach.

—

Where to find Adam Miller:

* LinkedIn: https://www.linkedin.com/in/adamjacobmiller/

* X: https://x.com/thatadammiller

—

In this episode, we cover:

00:00 Introduction to Marketplaces and Adam's Journey 02:15 The Evolution of LTV in Marketplaces 06:34 Key Metrics for Marketplace Success 08:44 Utilization Rates and Marketplace Dynamics 12:11 Building a Marketplace from Scratch 15:26 The Importance of Storytelling in New Categories 17:03 Common Threads Across Marketplace Industries 19:15 The Role of Marketing in Marketplace Growth 19:15 Marketplace Brand and Storytelling 23:42 Investing in Marketplaces: Key Considerations 25:58 Advice for Aspiring Growth Marketers 28:37 Testing Marketplace Ideas 30:30 Designing Effective Incentive Structures 33:15 AI's Role in Agentic Negotiation 33:15 Commodity Marketplaces vs Non-Fungible Marketplaces 35:11 Established Channel Expertise vs Emerging Experimentation 37:00 The T-Shaped Marketer 39:51 AI's impact on marketplaces 44:39 A day in the life of Adam Miller 50:12 The most common marketplace mistake 51:47 Lightning Round: Quickfire Questions

—

Obligatory disclaimer: I've worked at YouTube and Google for about a decade in various marketing teams. Nothing I say in my personal spaces is necessarily endorsed by them.

Listen now on YouTube, Spotify, and Apple.

—

Sundar Swaminathan is a growth marketing expert with 15 years of experience including at Uber, and is now a Marketing Data Science & Experimentation advisor to consumer tech scaleups and writes weekly in experiMENTAL.

Some takeaways:

1/ Magic Trip Metric = Your New BFF 🪄

Why it matters: Uber's genius "magic trip" metric quantified customer experience by breaking down each ride into measurable expectations vs. reality.

The real magic? It helped pinpoint exactly where experiences went wrong, making fixes surgical and effective.

The move: Create your own "magic" metrics by mapping your customer journey and setting quantifiable benchmarks for each step.

2/ Last Click Attribution?

Still Valid (Really!) 📱 Hot take: "Last click is wrong, but it's consistently wrong" - Sundar's refreshing perspective on keeping attribution simple.

For mid-stage companies with 2-3 channels, last-click can work just fine.

Pro tip: Focus on contribution over attribution. Test incrementality by holding one channel constant while varying another.

3/ The 90% Failure Club 🎯

Reality check: ~90% of experiments fail. But here's the twist - that's actually good!

The key is building a culture that celebrates learning from failures rather than avoiding them.

Power move: Don't sell experimental success; sell the process of generating 3-4 new ideas from each experiment.

4/ The PACE Framework for Data Leaders 📊

Break it down from Sundar’s course:

* Prioritization: 1 hour/week

* Analysis: 3-4 days max

* Communication: 1 full day

* Execution: 1-2 hours follow-up The secret sauce: Most analysts spend too much time on analysis and not enough on communication.

5/ The Anti-Service Model 🤝

Game-changer: Stop treating data teams as service providers.

Sundar's move? Weekly stakeholder-analyst meetings to establish partnerships, not support relationships.

Why it works: Better prioritization, more trust, and easier promotions.

6/ Data Democracy > Data Dictatorship 👑

Case study: Uber's Query Builder made SQL queries go viral (yes, really). Giving everyone access to data created a feedback loop of data obsession.

The lesson: Data tools don't create data culture - but they can amplify it.

7/ The Simple Stack Surprise 🛠

Plot twist: Uber's data science team primarily used just SQL and Google Sheets. Not Python. Not R. Just good ol' spreadsheets and queries.

The takeaway: Don't let tool FOMO hold you back. Focus on business acumen over technical complexity.

—

Where to find Sundar Swaminathan:

* Website: https://experimental.beehiiv.com/

* LinkedIn: https://www.linkedin.com/in/sswamina3/

—

In this episode, we cover:

00:00 Introduction 00:35 Uber's "Magic Trip" 03:46 Building a Data Obsessed Culture 05:12 Travis Kalanick's egoless leadership 07:59 7 Slides That Saved Uber $35M 11:22 When "extreme" experiments are justified 12:59 Why last click is sometimes all you need 16:34 Probabilities are being misunderstood 17:38 80-90% of experiments will fail 20:18 How many experiments are needed for a homerun? 21:33 Uber Eats started as an experiment 22:42 The experiment "success" trap 26:13 Building an experimentation strength in your company 27:49 You don't need fancy tools 29:30 How Uber made SQL internally viral 32:03 AI is not a panacea for marketing analysts 35:52 The PACE framework for data analysis 40:28 Where to invest your time as a marketing analyst 42:10 How to work effectively with analysts 44:16 The Journey into Newsletter Writing 48:21 Strategies for Growing Online Presence 49:51 How to grow your own following on LinkedIn 50:59 Future Aspirations and Projects

—

Obligatory disclaimer: I've worked at YouTube and Google for about a decade in various marketing teams. Nothing I say in my personal spaces is necessarily endorsed by them.

Listen now on YouTube, Spotify, and Apple.

—

Kevin Xu, perhaps better known by his Reddit moniker - Sir Jack, built an $8M fortune trading stocks on Reddit’s WallStreetBets before leaving his day job at YouTube to build AfterHour, a stock market super app. It raised $4.5M in it’s seed round last year.

Some takeaways:

🎯 Community Building 101:

* Go where your audience already is - Kevin saw 10x better results from engaging on Reddit vs traditional media like Fox Business or documentaries

* Authenticity trumps strategy - Kevin's SirJack persona succeeded because it wasn't originally meant to be anything, just genuine trading updates

* Communities need a clear reason to exist beyond just "having a community" - for AfterHour, it's helping people make money together

💡 Growth in 2025:

* Product differentiation is crucial - Kevin identified a key gap in finance platforms around trust and verification of trades

* Three viable paths to growth: novelty (viral AI features), authenticity (genuine creator connections), or enabling creators to monetize

* Mobile-first matters - Kevin intentionally built AfterHour as a "bathroom app" to capture daily habits and leverage push notifications

🎨 Design Philosophy:

* Ruthlessly eliminate complexity - AfterHour limits screens to 2 main buttons max

* Only show what matters most - resist the urge to display every possible metric just because you can

* Design for the average user, not power users - most people don't need every feature from Yahoo Finance

🚀 Platform Building:

* Communities need to evolve into "communities of communities" to achieve long-term sustainability

* Focus on unique value-add vs replicating existing tools - AfterHour emphasizes social features over duplicating stock data available elsewhere

* Trust and verification are crucial - Kevin built features to verify trades because screenshots alone weren't enough

🔑 Key Leadership Insight:

* Different phases need different skills - early stage (0-0.1) requires hands-on community building and DMing users individually, while scaling requires shifting focus to product and broader marketing initiatives

—

Where to find Kevin Xu:

* Website: https://afterhour.com/

* LinkedIn: https://www.linkedin.com/in/imkevinxu/

* X: https://x.com/imkevinxu

—

In this episode, we cover:

00:00 From Reddit Fame to Startup Success 01:29 The Journey of Entrepreneurship 04:20 The Rise of Wall Street Bets 06:23 Building a Community in Finance 09:34 The Transition from Anonymity to Identity 12:28 Creating a Financial Super App 15:14 Navigating Fame and Community Engagement 17:37 The Role of Authenticity in Community Building 20:42 Balancing Personal Brand and Business Goals 23:11 The Impact of Media on Business Growth 32:02 Building Community on Reddit 35:42 Onboarding and User Experience 37:20 Marketing Strategies on Reddit 39:34 The Role of Community in Marketing 41:24 Control Over Algorithms and Content 42:56 Algorithm Development for Diverse Users 43:45 Community Building Strategies 47:00 Sustaining Community Longevity 47:58 Mobile-Only Strategy 50:09 Design Principles for User Engagement 54:42 Leveraging AI in Development 57:38 Essentials for Building a Community 01:01:38 The Future of AfterHour and Community Dynamics

—

Obligatory disclaimer: I've worked at YouTube and Google for about a decade in various marketing teams. Nothing I say in my personal spaces is necessarily endorsed by them.

Subscribe to Horizons Podcast: https://horizonspod.com/

In this very short intro episode, Nate introduces Horizons Pod: a podcast exploring behind the scenes with marketers across the industry.

Be sure to subscribe to be notified when we launch soon!