Sprints are dead for prototyping
Product consultant & builder
IDEO had a line I always liked: if a picture is worth a thousand words, a prototype is worth a thousand meetings. For twenty years that was aspirational — prototypes were expensive, so we held the meetings.
It flipped. A working prototype is now quicker to build than the hour-long meeting you would book to discuss building it. I can put three clickable variants in front of a stakeholder in the time it takes to write the agenda. That kills the sprint as a prototyping tool.
The hour-long meeting vs the twenty-minute prototype
Think about what a refinement meeting for a new screen actually does. Eight people, one hour, and the output is a shared imagination of a thing none of them has seen. Then a sprint to build it, a review to discover the imagination was wrong, and another cycle to correct it. The ceremony was rational when building was the expensive part. It protected scarce engineering hours from bad guesses.
Building stopped being the expensive part. When the mockup costs twenty minutes, the meeting about the mockup is the waste. Reacting beats describing: put the real thing in front of people and the conversation collapses from an hour of abstraction to five minutes of "the second one, but move the price up."
What changed
Andrej Karpathy gave the shift its name in February 2025: vibe coding — talking software into existence and, in his words, forgetting that the code even exists. He built a working iOS app in an hour, in a language he was new to. The tweet was a throwaway; the capability underneath it was real, and everyone building with agents recognised it immediately.
My own version of that evidence is in this site’s build logs. 8tot8 got its entire look rebuilt in a single day of sixty small commits — palette, fonts, layout, copy, each one a decision made by reacting to a live page. Moo-ving shipped as a whole marketplace in its first commit. The prototype phase of those products happened at conversation speed.
The new loop
The loop that replaces sprint-based prototyping is simple: build three variants, put them in front of a real person, keep one, repeat. Decisions get made on reactions to real interfaces with real data in them. The role of the product person shifts from writing descriptions of things toward judging things — taste and prioritisation over specification.
This rewards a different meeting too. The useful session is thirty minutes around a live prototype, deciding. The wasteful session is sixty minutes around a document, imagining. I run engagements on the first kind.
Where sprints keep their job
Sprints still earn their keep where the constraint is coordination: many people, dependencies, integration risk, committed dates. Delivery at scale needs cadence and predictability, and a sprint provides both. The claim here is narrower and sharper: prototyping has left the sprint. Discovery now moves at build speed, and any process that makes a prototype wait two weeks for a ceremony is burning the cheapest learning available.
The takeaway
Prototypes went from expensive to nearly free, so the meeting became the costly artefact. Build the variants, react to real things, decide fast. Keep sprints for coordinating delivery — and keep them away from discovery.
My time is for hire.
Two ways to work with me: a full project build, where I design, build, and ship your product with these methods, or consulting inside your team.
Keep reading
The MVP is the easy part now
I have shipped five MVPs since February. Building them was the quick bit. Technology stopped being a moat, users are settling into fewer apps, and audience is the asset that decides who wins.
Big-company ideas, one-person company
Portfolio strategy, platform teams, ops reviews — the machinery of a €100M product organisation, scaled down to a company of one. Most of it fits, and agents made it affordable.