Aakash GuptaHow You Need to Think About Product Strategy in the Age of AI
At a glance
WHAT IT’S REALLY ABOUT
AI makes features cheap; strategy choices and metrics matter more
- The conversation argues that as AI makes feature delivery dramatically cheaper, product strategy becomes more critical because wrong choices now scale faster and cost more in adoption, maintenance, and distraction.
- Tim Herbig defines effective strategy as a set of opinionated choices that help teams say “no” over a time horizon, rather than a deck, template, or a long list of priorities.
- The episode introduces a practical stress test for strategy—the Roger Martin ‘opposite is stupid’ check—plus a ‘stupidity flip’ exercise to turn vague statements into real trade-offs.
- Herbig emphasizes that AI alone is not a defensible value proposition because models are broadly available; enduring differentiation more often comes from brand, data, network effects, and other durable ‘powers.’
- Using a fictional used-car marketplace (‘Fair Mile’), the hosts show how to narrow audiences with Audience Trees, choose winning moves, and build a strategy→metrics loop with measures directly tied to strategic choices rather than generic KPIs or machine metrics.
IDEAS WORTH REMEMBERING
5 ideasCheap features make bad prioritization more expensive, not less.
When it becomes easy to ship “something,” the real cost shifts to adoption, maintenance, and opportunity cost—getting the wrong choices into customers’ hands is what becomes expensive. Strategy’s job is to constrain output toward what changes customer and business outcomes.
The primary test of strategy is whether it helps you say “no.”
A strategy is only useful if it creates constraints that let teams reject plausible opportunities for the next 6–18 months. If your “strategy” is compatible with almost any roadmap item, it’s not doing strategy work.
Pressure-test strategy with Roger Martin’s ‘opposite is stupid’ test (the ‘stupidity flip’).
If the opposite of your stated strategy is obviously stupid, you haven’t made a real choice (e.g., “make salespeople more efficient”). Use the “stupidity flip” on specific components—audience, value prop, channel, differentiation—until the alternatives are believable.
AI is a resource, not a value proposition or moat by itself.
Because frontier models are widely accessible, “we use AI” is rarely a differentiator; it’s an input you leverage to amplify an existing customer value exchange. Sustainable advantage tends to come from slower-to-build assets like brand, data, and network effects.
Use Audience Trees to pick a focused segment, then use AI to size and sanity-check it fast.
Audience Trees help teams move from too-broad targets (e.g., ‘all car buyers in Europe’) to focused, defendable segments without becoming ‘comically narrow.’ AI can accelerate the market research and sizing needed to validate whether a narrowed segment is viable.
WORDS WORTH SAVING
5 quotesThese days, as everyone probably know, like producing features got kind of cheap, especially for greenfield products, but even established companies can get stuff out the door much, much faster. But the problem, the downside of that is when features get cheap, not having chosen the right feature, funnily enough, gets more expensive because suddenly you're flooding your product with all this output, all this fluffy stuff, all these great vibe-coded lovable things, and everything you do to circumvent the complex engineering work or even accelerated your engineering department. But this can come as a cost because you still have to make people use the features you produced.
— Tim Herbig
Whatever you're thinking your strategy is, whatever shiny deck it is, whatever one-pager it is, whatever framework, think about does it help me say no to things? That's all that matters.
— Tim Herbig
AI by itself in your strategy is not a value proposition. Everyone has access to the same frontier models. You can't compete on that technology alone.
— Tim Herbig
Well, the question is, is the opposite of your choice, of your strategy choice, stupid on its face? If the opposite is stupid, it's-- you haven't made a real choice.
— Tim Herbig
The point is, that's no value exchange. Like tokens used is not a value exchange for your customers. It's a value exchange for Anthropic.
— Tim Herbig
High quality AI-generated summary created from speaker-labeled transcript.
