RevOps Lab

Ravi Parimi has spent 20+ years in revenue operations and GTM strategy, from AWS, VMware and Cisco to Zendesk and ServiceTitan, plus a stint as executive in residence advising a16z portfolio companies. He joins Janis and Philipp to explain why the agent-first mindset breaks in production and why workflows, not agents, are the unit of design. Together they dig into human-agent handoffs, permission scopes and guardrails, and why RevOps has to run agents like a product after launch. Ravi closes with the unpopular opinion that most teams badly underestimate the operating model change ahead.

We cover:

- Why an agent-first mindset works for experimenting but fails in production

- Workflow as the unit of design: inputs, outputs, people and systems

- Lead routing as the test case: account disputes, strategic accounts and other edge cases

- Why deterministic rules matter if you want agents without hallucinations

- Treating agents as digital workers in the talent mix: what agents do best vs. humans

- Autonomy with a scope, like a driver's license

- How agent speed changes your org design and the humans in the loop

- Running agents like a product: roadmap, workflow v1/v2/v3, benchmarks and release cycles

- Agent ops: owning speed, quality, cost and latency after launch

- Why agents need sandboxes, testing and security layers before production

- Why modern revenue organizations will run as systems of agents and humans

- Why solution and field engineering leaders could be tomorrow's CROs

- Why judgment and taste become the premium skill when agents give the answers

Ravi Parimi: https://www.linkedin.com/in/rparimi/ 

Janis Zech: https://www.linkedin.com/in/janiszech/ 

Philipp Stelzer: https://www.linkedin.com/in/philippstelzer/ 

WeFlow: https://www.weflow.ai/ 

Community: https://www.weflow.ai/community 

RevOps Letter: https://www.weflow.ai/revops-letter 

Chapters:

(00:03) Ravi Parimi's path through RevOps

(02:59) What's wrong with agent-first

(06:54) Jobs to be done and lead routing

(09:56) Why the workflow mindset is missing

(13:54) Agents as digital workers

(15:44) Autonomy, scope and permissions

(17:13) Rethinking workflows and org design

(18:58) Running agents like a product

(21:11) Testing, sandboxes and guardrails

(24:13) The future of RevOps and the CRO

(28:48) Field engineers as future CROs

(32:01) Judgment, taste and revenue architecture

(34:26) Unpopular opinion: operating model

Janis Zech, CEO and founder of WeFlow, joins Kelly Lewis and Tiffany Jones on The Revenue Insiders to explain why AI in go-to-market fails without a clean data foundation. He walks through automated activity capture, AI meeting prep and coaching scorecards on MEDDIC/MEDDPICC, and why accurate forecasts depend on deal intelligence. The hosts push back on critical thinking, business context and vibe coding. Janis leaves listeners with one question for annual planning: are you AI-first or problem-first?

We cover:

- The three problems behind weak AI results in go-to-market: siloed data, non-selling time and unpredictable quota attainment

- Why manual data entry doesn't work and what to automate instead

- Design principles for data capture: integrate the signal systems, map everything back to the CRM, stay headless

- Why activity has to be mapped to opportunities, not just accounts

- AI meeting prep that cuts 15 minutes to 5 and lives in the calendar, Slack or email

- Coaching scorecards that auto-populate MEDDIC/MEDDPICC from every email, call and deal room

- Automation vs. human in the loop: why reps still need to understand the framework

- Why enablement and critical thinking matter more once AI does the data entry

- From system of record to system of intelligence to engagement orchestration

- Why AI forecasting failed without data pipelines, and how to give CROs multiple forecast numbers

- Feeding AI your business context: personas, industries, buying cycles

- Mining transcripts and emails for the real close-lost reasons

- Why vibe-coding your own GTM stack is slower and costlier than it looks

Original episode: https://www.therevenueinsiders.com/podcast/episode/39359e05/ai-first-or-problem-first-the-revops-question-for-2027 

Kelly Lewis: https://www.linkedin.com/in/kellylewis418 

Tiffany Jones: https://www.linkedin.com/in/tiffanyelizabethjones

The Revenue Insiders: https://www.therevenueinsiders.com 

Janis Zech: https://www.linkedin.com/in/janiszech 

Philipp Stelzer: https://www.linkedin.com/in/philippstelzer 

WeFlow: https://www.weflow.ai 

Community: https://www.weflow.ai/community 

RevOps Letter: https://www.weflow.ai/revops-letter 

Chapters:

(00:01) Janis's path to WeFlow

(03:43) Three problems in go-to-market

(05:48) What's different from Gong and Clari

(07:08) Design principles for data capture

(12:55) AI coaching on MEDDPICC

(15:40) Automation vs. human in the loop

(19:45) Enablement in an AI world

(23:34) System of intelligence

(24:52) Forecasting needs deal intelligence

(26:48) Feeding AI your business context

(30:58) Mining transcripts for real answers

(32:14) Vibe coding vs. long-term partnership

(35:07) AI-first or problem-first?

Avtar Varma, who just started as VP RevOps at Chainguard after building RevOps functions from Series A startups to Stripe, lays out three stages of enterprise AI adoption: token maxing, centralizing within functions, and a company-wide AI ops team. He explains why usage leaderboards are a vanity metric and what to reward instead. Together with Janis and Philipp, he debates where an AI ops function should sit, who owns the data infrastructure, and why go-to-market is the best place to prove ROI. He closes with an unpopular take on go-to-market engineers.

We cover:

- Stage one: token maxing, and why "give everyone every AI tool" makes ROI hard to prove

- Why AI usage leaderboards are a vanity metric

- Hackathons vs. leaderboards: rewarding the best ideas, not the biggest spend

- Meeting prep as a simple, high-impact use case, and why AI has to live in the calendar

- Stage two: centralizing skills, templates and agents inside each function

- Using usage data to coach below-average adopters instead of rewarding top spenders

- Stage three: a central AI ops team with a liaison for each function

- Where AI ops should sit: COO, CIO or RevOps, and who owns data infrastructure

- Saving cost with open-source models where best-in-class models aren't needed

- Why the data warehouse is becoming the layer agents run on

- Whether RevOps can grow into BizOps or AI ops

- Why go-to-market is the best place to start proving AI ROI

- Unpopular opinion: go-to-market engineers should be actual engineers

Avtar Varma: https://www.linkedin.com/in/avtar-varma

Chainguard: https://www.chainguard.dev/

Janis Zech: https://www.linkedin.com/in/janiszech/

Philipp Stelzer: https://www.linkedin.com/in/philippstelzer/

WeFlow: https://www.weflow.ai/

Community: https://www.weflow.ai/community

RevOps Letter: https://www.weflow.ai/revops-letter

Chapters:

(00:01) Avtar Varma joins from Chainguard

(01:58) Three stages of AI adoption

(05:54) Is token maxing necessary?

(07:33) Leaderboards as a vanity metric

(10:33) Hackathons and meeting prep

(14:46) Incentives that actually work

(16:48) Coaching with usage data

(18:10) Stage three: central AI ops

(21:07) Who owns data and infrastructure

(22:17) Cost, open-source models, security

(24:46) Can RevOps become AI ops?

(27:46) Why GTM is the place to start

(28:31) Unpopular opinion: GTM engineers

Colin Gerber, VP RevOps & Strategy at Socure, returns for part two to unpack how his team forecasts revenue in a fully consumption-based business. He walks through the solution readiness document that substantiates every booking, the "BAR" (booked ARR realization) process that tracks ramp against reality, and why RevOps should own the forecasting system but not its accuracy. Plus: the comp structures that keep AEs and CSMs honest, and the two mistakes that blow the biggest holes in a consumption forecast.

We cover:

- How Socure prices ~35 modules by API-call consumption, not seats

- Giving customers real-time visibility into what they're actually spending

- The solution readiness document that turns a sales pitch into a forecastable number

- Why AEs still call a bookings number — and when it gets challenged

- Booking conservatively for pre-launch startups vs. mature, high-volume customers

- Modeling seasonality into ramp schedules (BNPL, gaming, tax season)

- Using live trials to close the gap between closed-won and go-live

- The monthly BAR review with FP&A

- Salesforce, JIRA, NetSuite and Tableau: the four-source forecasting stack

- Booking discipline vs. realization — why bookings still matter in consumption pricing

- Who actually owns the ramp: AEs, CSMs, and post-sales solution consultants

- How 24-month payout schedules shape AE behavior on ramping accounts

- The two mistakes that blow up a consumption-based forecast

Colin Gerber: https://www.linkedin.com/in/colinsgerber/

Socure: https://www.socure.com/

Janis Zech: https://www.linkedin.com/in/janiszech/

Philipp Stelzer: https://www.linkedin.com/in/philippstelzer/

WeFlow: https://www.weflow.ai/

Community: https://www.weflow.ai/community

RevOps Letter: https://www.weflow.ai/revops-letter

Chapters:

(00:01) Colin Gerber returns from Socure

(02:37) How Socure prices consumption

(06:08) Real-time visibility into spend

(09:07) Why consumption pricing is spreading

(10:49) Forecasting sessions & the solution readiness doc

(15:07) Contract terms and seasonality

(17:41) Live trials close the go-live gap

(20:05) Ramp realization and revenue recognition

(21:55) The four-source forecasting stack

(25:05) Booking discipline vs. realization

(29:14) Who owns the ramp

(34:41) Two mistakes that wreck a forecast

(39:07) RevOps shouldn't own forecast accuracy

Nicole Bradshaw spent six months as interim global sales leader at Eventbrite — and discovered how much she didn't know about her own job. She joins Janis to explain why RevOps teams know exactly what good selling looks like but never apply it to themselves, and why an order-taking RevOps function is the one most easily automated away.

We cover:

• What changed when Nicole became her own stakeholder

• Why execs only want your top two problems, not all of them

• Balancing business impact against time to impact

• Using OKRs as a vehicle, not a solution

• Matching operating cadence to stakeholder level

• Strong opinions held loosely — showing up with a point of view

• Why order-taking RevOps is the easiest role to automate

• Bridging strategy and execution as RevOps' unique position

• Multi-threading inside your own company

• Picking the right champion for the right initiative

• Objection handling: anticipate, prepare, bring data

• Making people's lives easier before you need favors

• Framing every ask as a win-win

Nicole Bradshaw on LinkedIn: https://www.linkedin.com/in/nkbradshaw/ PandaDoc: https://www.pandadoc.com Janis Zech on LinkedIn: https://www.linkedin.com/in/janiszech/ Philipp Stelzer on LinkedIn: https://www.linkedin.com/in/philippstelzer/ WeFlow: https://www.weflow.ai/ Join the RevOps Chat Community: https://www.weflow.ai/community Subscribe to the RevOps Letter: https://www.weflow.ai/revops-letter

Resource recommendation: Harvard Business Review — Nicole's habit of blocking 30 minutes every other week for professional development reading

Chapters: (00:00) Intro & welcome to Nicole (00:37) Nicole's path into RevOps (03:08) Six months leading sales (04:43) Becoming your own stakeholder (07:34) Impact vs. time to impact (09:28) OKRs and operating cadence (13:08) Showing up with agency (17:19) AI-proofing the RevOps role (18:26) Bridging strategy and execution (22:56) Multi-threading and champions

Dan Jiao, VP RevOps at Aircall, has now moved two companies from a global functional model to regional GMs — at Signifyd and again at Aircall. He joins Janis to explain what actually changes when four business units each own their P&L, and why RevOps becomes the glue that keeps one company from quietly becoming five.

We cover:

• Global functional vs. regional GM model — and what stays centralized
• The revenue scale where a regional split starts to make sense
• Running RevOps as a product org: the global go-to-market operating system
• Who owns segmentation vs. who co-designs territories
• Embedding regional business partners as de facto chiefs of staff
• One global scoring model, multiple regional calibrations
• Why capacity ratios differ market by market
• RevOps as the neutral referee in budget and headcount debates
• Holding one global forecasting standard across four business units
• Buyer-verified exit criteria instead of rep-filled fields
• Sharing best practices when there's no global head of sales

Dan Jiao on LinkedIn: https://www.linkedin.com/in/danjiao/
Aircall: https://aircall.io
Janis Zech on LinkedIn: https://www.linkedin.com/in/janiszech/Philipp Stelzer on LinkedIn: https://www.linkedin.com/in/philippstelzer/WeFlow: https://www.weflow.ai/Join the RevOps Chat Community: https://www.weflow.ai/community
Subscribe to the RevOps Letter: https://www.weflow.ai/revops-letter

Book recommendation: Switch: How to Change Things When Change Is Hard by Chip & Dan Heath

Chapters:
(00:00) Intro & welcome to Dan(02:33) Dan's path from sales to RevOps(04:46) Global functional vs. regional GM(08:12) When a regional split makes sense(11:18) Who owns segmentation and territories(14:04) The hybrid model and regional business partners(17:43) Account prioritization and scoring(20:49) Capacity planning across regions(25:53) RevOps at the budget table(27:52) One global forecasting standard(30:31) Buyer-verified exit criteria(32:43) Sharing best practices without a global sales lead(34:50) Book recommendation & close

Mollie Bodensteiner, VP RevOps at ZoomInfo and returning guest, joins Janis to break down how AI is reshaping RevOps roles, team structures, and career paths. RevOps used to report on the business — now it builds and governs the systems that run it. Mollie shares what the go-to-market engineer role actually is, why career ladders are turning into portfolios, and why she stopped doing case studies in hiring.

We cover:

• From reactive help desk to building and governing the systems that run the business

• AI sprawl: why "give everyone access" now creates a governance problem

• What a go-to-market engineer really is — beyond the LinkedIn hype

• Agent ops: monitoring drift, QA, and cost like DevOps monitors uptime

• Why it's a skill set, not a new job title

• The analyst shift from firefighting to proactive intelligence

• Who should own AI: centralized, federated, or center of excellence

• Flatter orgs, wider spans, and horizontal layers replacing functional silos

• Career paths becoming portfolios instead of ladders

• Why Mollie killed case studies in her hiring process

• Skill vs. will — advice for early-career operators

Mollie Bodensteiner on LinkedIn: https://www.linkedin.com/in/molliebodensteiner/ ZoomInfo: https://www.zoominfo.com Janis Zech on LinkedIn: https://www.linkedin.com/in/janiszech/ Philipp Stelzer on LinkedIn: https://www.linkedin.com/in/philippstelzer/ WeFlow: https://www.weflow.ai/ Join the RevOps Chat Community: https://www.weflow.ai/community Subscribe to the RevOps Letter: https://www.weflow.ai/revops-letter

Book recommendation: The Accountable Organization by John Marchica

Chapters:

(00:00) Intro & welcome back to Mollie

(03:26) How RevOps jobs have changed

(05:30) AI sprawl and the governance problem

(08:43) The go-to-market engineer role

(12:06) Agent ops and the digital workforce

(16:25) What happens to the analyst

(20:30) Who owns AI in the org?

(24:19) Flatter teams, wider spans

(28:20) Making RevOps impact measurable

(31:43) Why case studies are dead in hiring

(34:11) Skill vs. will: advice for operators

(37:07) Book recommendation & close

Colin Gerber, VP RevOps & Strategy at Socure, breaks down 17 years of CPQ evolution — from Zuora's billing-wrapper days to today's consumption-ready, platform-agnostic tools. He shares how Socure pressure-tested its pricing in Excel for a year before building anything, and how a new entitlements object now ties CRM, CPQ, and billing into one source of truth.

We cover:

• CPQ's three eras: Zuora → Salesforce CPQ → platform-agnostic tools like DealHub

• Pressure-testing pricing in Excel before implementing CPQ

• Socure's shift to à la carte, consumption-based pricing across 35 modules

• Common CPQ mistakes: over-complication and workflow bloat

• Scaling SKUs across ~170 countries

• RevOps' role: guardrails, deal desk, pre-deal margin modeling

• Using AI to auto-generate Solution Readiness Documents

• A bi-directional entitlements object unifying CRM, CPQ & billing

• Automated entitlement capture for PLG motions

• A preview of consumption-based forecasting

Colin Gerber on LinkedIn: https://www.linkedin.com/in/colinsgerber/ Socure: https://www.socure.com Weflow: https://www.weflow.ai/ RevOps Chat Community: https://www.weflow.ai/community RevOps Letter: https://www.weflow.ai/revops-letter Janis on LinkedIn: https://www.linkedin.com/in/janiszech/ Philipp on LinkedIn: https://www.linkedin.com/in/philippstelzer/

Book recommendation: Team of Teams: New Rules of Engagement for a Complex World by General Stanley McChrystalChapters: (00:00) Intro & welcome to Colin (01:51) Colin's background (04:23) The three phases of CPQ history (10:00) CPQ challenges with consumption-based pricing (14:12) Common CPQ implementation mistakes (19:03) The role of RevOps in CPQ (21:19) Using AI in CPQ (25:21) Where's the system of truth? (32:26) Direct vs. self-service/PLG (34:44) Teaser: consumption-based forecasting (36:44) Book recommendation & close

Markus Jaensch, Head of RevOps at Aiven, joins Janis and Philipp to unpack what consumption-based pricing actually means for a RevOps team running a $100M+ ARR business. Aiven — the open-source data platform behind managed Kafka, Postgres, OpenSearch, and ClickHouse — recently crossed $100M ARR, and Markus walks through how the forecasting model, comp design, and territory setup have evolved over 4.5 years to deal with the fundamental problem of consumption: a "win" doesn't equal revenue, and the next 12 months can swing either direction.

We cover:

• Why consumption ARR is structurally harder to forecast than SaaS bookings
• Why Aiven moved away from pure ARR forecasting and back to bookings + finance-modeled ARR
• The three forecast buckets: new business, add-on, and commit contracts
• Using MEDDPICC with mandatory mutual success plans and CRM score as a red-flag signal
• T-shirt sizing (S/M/L/XL) with solution architects to anchor opportunity size
• The two non-negotiables to close-won at Aiven: valid payment method + 3 consecutive days of consumption
• Why RevOps owns the close-won gate — and the tradeoff of being a controlling function
• The Farmer/Hunter evolution: from split, to mixed, to dedicated Inside Sales + named-territory Field reps
• Comp design: bookings as a sanity metric, ARR as the paid metric, with new-ARR ramp-to-size
• Stack: Salesforce + data warehouse, and why Aiven switched from spot ARR to a 30-day average
• Building the process around the customer first, not internal convenience

Markus Jaensch on LinkedIn: https://www.linkedin.com/in/markus-jaensch/

Aiven: https://aiven.io

Weflow: https://www.weflow.ai

RevOps Chat Community: https://www.weflow.ai/community

RevOps Letter: https://www.weflow.ai/revopsletter

Janis on LinkedIn: https://www.linkedin.com/in/janiszech

Philipp on LinkedIn: https://www.linkedin.com/in/philippstelzer

Book recommendation: The Café on the Edge of the World by John Strelecky

Chapters:

• (00:00:00) Intro & Welcome to Markus
• (00:02:14) Why Consumption ARR Is the Topic — and Aiven Crossing $100M
• (00:03:39) Why Consumption ARR Is So Hard to Forecast
• (00:08:30) Bookings vs. ARR Forecasting — and Aiven's Evolution
• (00:12:53) MEDDPICC, Mutual Success Plans & Deal Hygiene
• (00:16:21) T-Shirt Sizing with Solution Architects
• (00:18:01) The 3-Day Consumption Rule for Close-Won
• (00:22:28) Modeling the ARR Ramp After a Booking
• (00:26:27) The Farmer/Hunter Evolution & Inside Sales Motion
• (00:29:37) Comp Design: Bookings, ARR, and Ramp-to-Size
• (00:31:00) Stack: Salesforce, Data Warehouse & 30-Day Average ARR
• (00:34:34) Final Learnings on Consumption Forecasting
• (00:36:29) The Customer-First Principle
• (00:38:02) Book Recommendation & Close

In this episode of the RevOps Lab, Janis sits down with Andy Smidmore — a 15-year RevOps and SalesOps veteran whose résumé reads like a tour of high-growth West Coast SaaS (Confluent, Cloudera, Docker, most recently Ditto). Andy is also the author of a popular LinkedIn article series on RevOps fundamentals, including the piece that anchors today's conversation: Forecasting Is a Trust Problem Before a Math Problem. Janis and Andy unpack why the magic forecast number leadership wants is just the surface, why broken sales stages quietly leak forecast accuracy, and why mixing deal reviews into your forecast call destroys the very trust your reps need to be honest with you.

We cover:

• Why forecasting is a trust problem first, a math problem second

• Why the "how much are you committing?" question on day one of a forecast call is the wrong starting point

• Building sales stages around the buyer's journey, not your selling process — and why misaligned stages are a leaking bucket

• The danger of performative deal progression and the false hope it creates for leadership

• Why deal reviews and forecast calls should be two separate motions

• "I'm not questioning you, I'm asking questions" — the subtle reframe that changes the whole room

• Crystal-clear definitions for pipeline vs. commit, kept short (5 bullets, not essays)

• What a true commit actually requires: technical sign-off, legal redlines, procurement engaged, stakeholder alignment

• Why a consistent week-in, week-out forecast format builds rep trust over time

• Treating forecasting as a team sport — and why punishing reps swings them toward concealment, not honesty

Links:

• Andy Smidmore on LinkedIn: https://uk.linkedin.com/in/andy-smidmore-3aab0556

• Andy's article series on RevOps & forecasting (LinkedIn)

• Janis Zech on LinkedIn: https://www.linkedin.com/in/janiszech/

• Philipp Stelzer on LinkedIn: https://www.linkedin.com/in/philippstelzer/

• WeFlow: https://www.weflow.ai/

• Join the RevOps Chat Community: https://www.weflow.ai/community

• Subscribe to the RevOps Letter: https://www.weflow.ai/revops-letter

• Book recommendations:

• Disrupted by Dan Lyons

• The Girl with Seven Names by Hyeonseo Lee

Chapters:

• (00:00) Intro & welcome to Andy

• (00:51) Andy's 15-year journey through Confluent, Cloudera, Docker, Ditto

• (03:08) Why forecasting is the topic — and the article that started it

• (04:03) Why every forecast framework is unique to its company

• (06:15) Intelligence tools haven't replaced the foundational framework

• (08:13) "How much are you committing?" — the surface-level forecast call

• (10:01) The top forecast process problems Andy keeps seeing

• (11:33) Sales stages built around the buyer journey, not the seller

• (12:24) The POC stage problem: what does "done" actually mean?

• (14:25) When poor stage definitions destroy trust between reps and leadership

• (15:43) Creating an open, honest forecast call environment

• (18:17) From gut-based reviews to data-backed conversations

• (20:22) Why deal reviews and forecast calls must be separate motions

• (23:25) "I'm not questioning you, I'm asking questions"

• (25:00) The cadence: clean pipeline Thursday, deal review Friday, forecast Monday

• (28:32) Building a clear forecasting methodology — pipeline vs. commit

• (31:34) What actually constitutes a commit deal

• (35:21) Closing thoughts: stages, enablement, cadence, culture

• (39:35) Book recommendations & close