Skip to content
Lenny's PodcastLenny's Podcast

Shaun Clowes: Why AI products win on data, not on the model

What if great AI product work is really data management work; Confluent's leader grounds PMs outside the building, where customer insight differentiates.

Lenny RachitskyhostShaun Clowesguest
Dec 28, 20241h 21mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Product, Data, And AI: Why Most PMs Still Miss The Point

  1. Shaun Clowes, CPO at Confluent, argues that product management remains an underdeveloped discipline where outcomes and PM quality are highly variable, largely because most PMs focus too much inside the building and not enough on customers, markets, and data.
  2. He lays out what distinguishes great PMs: structured, outside-in thinking, rigorous use of data as a compass (not a GPS), and the ability to synthesize diverse feedback into sharp, contrarian insights that drive real business value.
  3. Clowes explains why AI’s real leverage in product work is all about data management—getting the right, fresh, well-structured data into models—rather than clever prompts or UX tweaks, and how this will strengthen, not weaken, major SaaS incumbents.
  4. He also shares lessons from building one of the first B2B growth teams at Atlassian, the nuanced realities of PLG in B2B, his “career as bingo card” philosophy, and candid reflections on failed products and over-indexing on data.

IDEAS WORTH REMEMBERING

5 ideas

Great PMs start outside the building and stay there.

The highest-leverage PMs frame everything from the customer, market, and competitor perspective, rather than getting sucked into sprint management, internal politics, or delivery coordination. Documents, strategies, and discussions should all begin with external realities, not internal org charts.

Data is a compass, not a GPS—and most PMs misuse it.

Data rarely hands you the answer; it helps you falsify bad hypotheses and sanity-check intuition. Clowes advises going one step upstream, downstream, and one level up from any metric (who it affects, representativeness, long-term impact, revenue quality) before declaring a ‘win’ or making big bets.

AI’s real advantage in product work comes from data management, not clever prompting.

LLMs are powerful synthesis engines but “insanely dumb” without timely, comprehensive, well-structured data. The hardest and most valuable work—both for internal tooling and customer-facing AI features—is aggregating, cleaning, connecting, and continuously refreshing the right data to feed these models.

Enterprise SaaS moats are in business rules and configuration, not just UI.

Products like Salesforce, Workday, or Jira are not easily cloned because their real lock-in is years of accumulated, company-specific workflows and rules embedded in their systems. Even in an AI/agentic world, those rules still govern what agents can do, so incumbents may actually get stronger.

PLG rarely replaces sales; the winning model blends both motions.

Product-led growth is less about eliminating sales and more about having a system and incentives that prioritize end-user success early in the journey. The most resilient B2B companies use PLG to generate and qualify demand, and sales to monetize and deepen relationships, aiming for both many customers and high revenue.

WORDS WORTH SAVING

5 quotes

The job is not about execution; it’s about finding reliable, differentiated value you can uniquely deliver into the market.

— Shaun Clowes

LLMs are limitless information eaters, but they’re also insanely dumb. They only know what you train them on or what you give them right now.

— Shaun Clowes

People really underestimate where the value is created in these applications. It’s not the forms on databases; it’s the years of evolved business rules and workflows.

— Shaun Clowes

Data is more like a compass than a GPS. If you expect it to give you the answer, you’ll either be wrong or slow—often both.

— Shaun Clowes

People don’t care what you know until they know that you care.

— Shaun Clowes

Why most product managers underperform and what great PMs actually doUsing customer, market, and competitive insight as the core of product workAI’s impact on product management: data management vs. models and UXHow data should and shouldn’t guide product decisions (data as compass, not GPS)Building and scaling B2B product-led growth and its interplay with salesHow AI will (and won’t) disrupt enterprise SaaS and systems of recordCareer strategy for PMs: breadth, versatility, and “bingo card” experiences

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome