PLG to Enterprise GTM: RevOps Playbook for Experimentation, Signal Discipline, and AI That Actually Works A conversation with Stephanie Couzin. Executive Summary Lucid’s GTM evolution is not a “PLG vs Sales” story. It’s an operating model story. In this episode, Stephanie Couzin (VP, GTM Strategy & Ops at Lucid) breaks down how modern revenue teams can scale experimentation, build enterprise sales muscle, and adopt AI without turning their tech stack into a Frankenmonster. Readers will learn: PLG → Enterprise is a signal shift: you move from optimizing for users to optimizing for accounts, buying committees, and expansion readiness. Experimentation only works if comp risk is managed: GTM tests touch variable pay, so pilots must be designed as true win-wins. Psychological safety is a growth lever: teams ship better ideas faster when people can speak up, fail fast, and share learnings without fear. Data is the cost of entry for GTM testing: if upper-funnel metrics and activity data are messy, your “experiments” turn into opinions. Standardize AI or suffer whack-a-mole: pick a core AI platform to reduce tool sprawl and enable repeatable adoption. Agentic vs Copilot AI are different games: agentic is replacement (parity + cost savings), copilot is augmentation (productivity + more customer time). Start with the use case, not the tool: the fastest path to value is defining the workflow problem first, then deciding buy/build. One priority beats twelve “priorities”: focus drives execution, and execution is the only feature that matters. Facebook Twitter Youtube The Hidden Shift: From “Users” to “Accounts” Early Lucid was deeply product-led, built on a mature self-serve engine. Then came the layering: Sales motion Segmentation Enterprise complexity Multi-threaded post-sales workflows Stephanie describes it as moving from user-centric growth to account-centric expansion. PLG gives you usage.Enterprise GTM demands orchestration. The real challenge isn’t adding sales reps.It’s building the infrastructure to know when, where, and why sales engagement should happen. Experimentation Built Lucid’s Sales Muscle Stephanie points to Lucid’s early experimentation with a Product Qualified Motion. Most PLG companies start with lead scoring at the user level. Lucid evolved it further: Not just which user is engaged But what is the account telling us? And who should we reach out to now? That shift is everything. “Where that really evolved over time was identifying at the account level, what are the signals we need to outreach at the right time with the right messaging.” — Stephanie Couzin That’s the marriage of: First-party product signals Third-party intent and context Segment-aware targeting Persona-aware outreach Stephanie calls it the “special sauce” of modern lead scoring. The Account Expansion Checklist (Lucid’s Internal Recipe) Stephanie hints at what many GTM teams lack: an internal definition of “ready.” Lucid has an internal checklist: “A set of account signals where we say: this account is locked and ready to expand.” If the checklist isn’t met, Lucid doesn’t stop. They reverse-engineer value: You use Slack? You use documentation tools? You have workflows Lucid integrates into? Then Lucid doesn’t sell harder. They sell smarter: “This is how you get more value.” Cross-Functional Alignment: Experimentation Without Chaos Testing in GTM is not like testing in product. Because in revenue… someone’s commission is always involved. Stephanie puts it bluntly: “Testing something new will impact someone’s variable compensation.” — Stephanie Couzin So experimentation requires designing for buy-in. Lucid ran a coverage pilot by: Using lower-risk accounts Making rep trades fair Ensuring nobody felt punished for participating A GTM experiment only works if it’s a win-win. Otherwise reps sandbag, ignore, or resist. And your “pilot” becomes theater. Psychological Safety Is the Real GTM Scaling Lever Stephanie goes deep here, referencing Google’s Project Aristotle – Google studied what makes teams high-performing. The #1 factor wasn’t IQ.It wasn’t experience.It wasn’t process. It was: psychological safety. “They could speak up without fear of consequence. They felt freedom to fail fast.” — Stephanie Couzin Stephanie teaches this internally at Lucid. And she’s clear: This isn’t about politeness.It’s about intentional leadership. Practical mechanisms Lucid uses: Weekly project show-and-tells “Smart Fridays” where reps share plays Normalizing learnings over perfection The GTM Psychological Safety Loop Safe to speak → More ideas → Faster experiments → Better learning → Higher trust → Stronger execution Data Discipline: The Unsexy Requirement for GTM Experiments Stephanie delivers one of the hardest truths in RevOps: Lower funnel data is solid. Upper funnel data is… suspicious “As you move further up funnel… fewer eyes are on those metrics.” — Stephanie Couzin Testing requires: Full-funnel accuracy Integrated activity capture Shared KPI definitions Governance (tagging, segmentation discipline) AI in GTM: Standardize or Suffer Stephanie’s AI governance advice is refreshingly simple: “Standardize on something. Otherwise you’re in whack-a-mole.” Lucid standardized on Google Gemini. Not because it solves every revenue use case. But because: It reduces fragmentation It sparks shared experimentation It creates repeatable workflows It prevents reps from duct-taping random tools together The goal is not AI everywhere. The goal is AI that fits into systems. The Most Important GTM AI Rule: Start With Use Cases, Not Tools Stephanie nails this: “What use case are you trying to solve, not what tool do you want?” Because shiny object syndrome is real. Most orgs buy AI like toddlers choosing cereal: “Ooh, the box is shiny.” Lucid forces the opposite: Define workflow pain first.Then evaluate tooling. AI Use Case Intake Form What GTM workflow breaks today? What manual effort exists? What does “better” look like? Is this assistive or agentic? Can current stack solve 80% already? What data sensitivity is involved? What adoption friction will occur? Agentic AI vs Copilot AI: Stop Confusing the Two Agentic AI Replaces human work Measured in cost savings Goal is parity (“do no harm”) Copilot / Assistive AI Enhances workflows Measured in productivity and customer time Harder to quantify, but more transformative “Most AI we adopt for revenue teams is assistive. It makes teams more productive.” That’s where the real gains are: Better prep Faster follow-up Less manual CRM work More customer-facing time The GTM Takeaway Stephanie Couzin’s playbook isn’t about