Lenny's PodcastLessons on product sense, AI, the first mile experience, and the messy middle | Scott Belsky (Adobe)
CHAPTERS
- 0:00 – 1:14
Conviction as the north star: when to stick it out vs quit
Scott shares a decision framework he uses with founders whose companies are struggling: measure how conviction has changed since the start. If conviction has increased with new learning, you're likely in the "messy middle" and should persist; if conviction has dissipated, it's time to pivot or stop.
- •Ask: do you have more or less conviction now than when you started?
- •More conviction despite setbacks = messy middle; keep going
- •Less conviction with more information = quit or pivot; life is short
- •Conviction is the fuel required to endure volatility
- 1:14 – 7:12
Scott’s trajectory: Behance to Adobe, and what he does now
Lenny introduces Scott’s background, then they unpack Scott’s current Adobe role spanning strategy, corp dev, design, and emerging products. Scott describes how Adobe’s priorities shifted from cloud/collaboration to AI and new growth areas, changing the focus of his day-to-day work.
- •Behance founder acquired by Adobe; rose to CPO, then shifted roles
- •Current scope: company strategy, M&A/corp dev, design org, emerging products
- •Past chapter: cloud transition, collaboration, connectivity, acquisitions
- •New chapter: AI-driven opportunities, 3D/immersive, stock/media transformation
- 7:12 – 8:29
Why many CPOs don’t last—and Scott’s view on breaking org boundaries
They discuss why CPO tenures can be short and what helped Scott thrive at Adobe. Scott argues that product wins come from user experience, and he emphasizes collapsing artificial boundaries among product, design, and engineering.
- •CPO churn often stems from mismatched expectations and limited scope
- •Scott’s CPO model included design + product + engineering alignment
- •Function boundaries can be “artificial” and constrain outcomes
- •Experience of the technology often matters more than the technology itself
- 8:29 – 9:26
Building product sense: empathy over attachment to solutions
Scott explains that product sense begins with deep empathy for the customer rather than passion for a preconceived solution. He frames product craft as applied psychology—understanding human tendencies and designing with them in mind.
- •Teams over-index on solutions instead of understanding customer pain
- •Empathy often reveals the real solution (your idea may be 30° off)
- •Great product design is largely psychology and human tendencies
- •Product sense grows through repeated exposure to real customer context
- 9:26 – 13:22
The “first mile” experience: onboarding, orientation, and defaults
Scott defines the first mile as the initial seconds/minutes where users form judgments and either activate or churn. He explains why this phase is often neglected, how psychology shows users are “lazy, vain, and selfish” early on, and why improving the first mile can unlock major growth.
- •First mile = onboarding + orientation + defaults + knowing where you are
- •In first ~30 seconds, users want fast success and minimal learning
- •Teams often focus on the final mile of building, not the first mile of use
- •Getting more users through activation is a hallmark of world-class teams
- 13:22 – 16:34
How to develop empathy: go “shoulder to shoulder” with customers
Scott shares the most effective way he’s learned to build empathy: observe customers in the full context of their day, not just during product usage. Seeing surrounding tools, interruptions, and constraints builds intuition that surveys and dashboards miss.
- •Observe customers in their real environment, not just scripted tests
- •Context (meetings, other apps, family life) shapes product usage
- •This exposure creates better intuition and product decision-making
- •Empathy lets makers “be the customer,” enabling more authentic design
- 16:34 – 20:42
Consumer products that endure: network effects, insights, and platform risk
Scott contrasts durable consumer successes (Uber, Pinterest) with “flash in the pan” apps whose novelty gets copied by incumbents. He highlights the importance of underlying structural advantages—network effects, untapped capacity, and unique psychology insights—beyond clever interfaces.
- •Durable winners often build new network effects (e.g., Uber)
- •Unique consumer psychology insight can create staying power (e.g., Pinterest)
- •Structural distribution loops matter (pins driving traffic back to sites)
- •Novel interfaces alone are easy for platforms to copy and absorb
- 20:42 – 26:15
“Only do half”: ruthless reduction and optimizing for the right problems
Scott explains his philosophy that teams should build fewer features, offer fewer options, and focus on a narrower wedge. Using Behance as a case study, he shows how removing features increased focus on the core metric and improved product clarity, often with minimal long-term backlash.
- •MVP guidance: optimize for the problems you want to have later
- •Remove brick walls; don’t pre-build every future request
- •Behance launched too complex; killing features boosted the core metric
- •When adding, ask what you can replace or remove to reduce cognitive load
- 26:15 – 29:40
AI’s 5-year impact: from workflow to flow, and the rise of experiences
They shift into AI, and Scott shares an optimistic view: AI reduces repetitive labor and friction, unlocking human ingenuity. He argues productivity gains may increase ambition rather than reduce headcount needs, and he sees an expanding “experience economy.”
- •AI moves us from “workflow to flow” by removing mundane friction
- •Higher productivity can increase demand/ambition, not shrink teams
- •Humans find edges; AI often regurgitates the center (still valuable)
- •More time for non-scalable, high-touch experience improvements
- 29:40 – 32:43
How AI changes product teams: collapsing the stack and empowering roles
Scott predicts AI will reduce “operator” bottlenecks by letting people answer their own questions across data, design, and execution. This flattens organizations and helps larger companies behave more like small, fast teams.
- •AI enables people to self-serve insights instead of routing requests
- •Less “game of telephone” across functions and layers
- •Big orgs can regain some speed advantages of small teams
- •Specialists shift from repetitive tasks to higher-leverage work
- 32:43 – 35:18
How PM work evolves: expanding the surface area of possibilities
Scott describes top performers as those who preserve time to explore many options, then narrow through feedback and refinement. AI’s superpower is accelerating exploration—helping teams see more alternatives (including what not to do) faster.
- •Great builders explore broadly, then cull and refine to a few options
- •AI expands ideation and scenario exploration at low cost/time
- •Value includes revealing “anti-options” (what you shouldn’t do)
- •PMs gain leverage by spending more time on judgment and curation
- 35:18 – 38:13
Design upstream, “golden gut,” and how AI augments creative iteration
Scott advocates bringing design earlier into product development so designers develop intuition from customer and product debates. He defines “golden gut” as experienced instinct for simplifying flows and reducing cognitive load, and he outlines how generative AI can accelerate iterative creative leaps.
- •Bring designers into early customer research and value-prop debates
- •“Golden gut” = micro-instincts that simplify, default smartly, and reduce load
- •AI enables real-time generative options (e.g., extend/outpaint in Photoshop)
- •Future: more leap-based creativity via AI-assisted next-step suggestions
- 38:13 – 41:49
Staying ahead of AI: play, write, and create forcing functions
Scott’s advice for PMs is simple: play with the tools to understand what’s possible, and follow curiosities. He describes his Implications newsletter as a forcing function to track change and deepen thinking, then shares practical discipline for writing amid a busy schedule.
- •Core advice: “Play” with AI tools regularly to avoid getting stuck
- •Pursue curiosities; experimentation builds intuition about capabilities
- •Writing (e.g., Implications) forces synthesis and long-term idea capture
- •No hack for writing: ruthless time prioritization and small daily windows
- 41:49 – 47:03
The messy middle: endurance, progress narratives, and knowing when to quit
Scott explains why the middle is volatile and psychologically difficult: humans crave constant rewards, but transformative work takes years. He recommends manufacturing micro-milestones and “merchandising progress,” then returns to his conviction test as the best lens for deciding whether to persist or pivot.
- •Messy middle is full of lows: anonymity, uncertainty, anxiety
- •Use micro-goals and celebrations to short-circuit the reward system
- •Leaders must narrate progress to avoid teams going “stir crazy”
- •Quit/pivot when conviction meaningfully declines (but don’t decide on one bad day)
- 47:03 – 1:02:32
Angel investing and leadership principles: truth-tellers, object models, resourcefulness
Scott shares what he looks for in founders (listeners, mission-driven, pragmatic truth-tellers) and in products (clear object models and navigational clarity). He closes with broader guidance: in constrained times, resourcefulness beats resources, and extraordinary outcomes require non-ordinary approaches.
- •Founder traits: listening, learning, mission > money, pragmatic about obstacles
- •Avoid promoters who sugarcoat; prefer optimism about future + realism about present
- •Product lens: strong object model—“How did I get here? What now? What next?”
- •Resourcefulness > resources in downturns; refactor systems instead of throwing headcount
