Shape Up: How to Ship Without Losing Your Startup Soul
Ryan Singer
Creator, Shape Up · Former Head of Product, Basecamp
2019–2025
The Core
Three Constraints That Create Shape Up
"We're not going to take a big concept and ask 'What's the estimate?' We're going to ask 'What's the maximum amount of time we're willing to spend?' and come up with an idea that works in that time."
David (Rails creator) was coding only 10 hours/week at the start — scarcity forced efficiency
Jason's obsession with forward momentum meant no tolerance for fuzzy, sprawling projects
Intense collaborative shaping sessions (Jason + Ryan with Sharpie pen on paper) replaced waterfall specs
Three-hour working sessions, all stakeholders present, trying to break ideas before building
What Shape Up Actually Is
A Different Way to Think About Building
6
week cycles
3
hour shaping sessions
1
team with no silos
Appetite: Not estimate, but "How much time can we spend on this?" (e.g., 6 weeks)
Framing: Narrow the problem down. What's really worth solving?
Shaping: Three hours minimum with Product + Design + Engineering. Try ideas, break them fast, de-risk before you ship
Build: With clarity. Ship what's possible in the time. No "this is too complicated," figure out tradeoffs upfront
Why this mattersYou're not discovering that the task is impossible in week 4. You're discovering it in the shaping room before anyone is resourced.
The Shaping Session
How to De-Risk Before You Build
Invite: Product + Design + At least one engineer (the "grumpy old plumber" who insists on looking at the pipes)
Time: Three hours initial, then maybe break and come back. Fast working, not async review cycles
Process: Try different approaches. Not one idea for three hours. Step back and ask, "What if we did this a completely different way?"
Goal: See the end from the beginning. Every person in the room should walk out saying, "I know what this is and I can build/design/ship it."
The FinTech Example
Team thought they'd reduce onboarding by piping in bank data. But the code had three different branches per bank. Found in shaping = easy conversation. Found in week 4 = project blowup.
De-Risk the Rabbit Holes
Technical person opens the code. Designer plays through the customer scenario. Product looks at edge cases. These "what-ifs" are not details. They're the difference between shipping and failing.
Integration
The Shaping Session Works When...
All three disciplines (product, design, engineering) are in the room at the same time
The engineer is there to say "that won't work" early, not validate it later
You have permission to say "I don't know" and actually explore instead of guess
You stop going in circles on one idea and try completely different approaches
The human elementPeople aren't used to working this fast. You need to face the blank page together. No documents to comment on tomorrow. Just real-time, face-to-face problem-solving.
Contrarian
The Shape Up Truths Most Teams Get Wrong
✗Estimate the project, then build itINSTEAD →✓ Set an appetite (time you're willing to spend), then fit the idea into that appetite
✗Design in Figma, then show engineersINSTEAD →✓ Engineer in the room while you're designing. They catch the time bombs now
✗Ship big features on 2-week sprintsINSTEAD →✓ Ship finished, shippable work in 6-week cycles. Longer cycles = more meaningful stuff
✗Small companies scale by adding processINSTEAD →✓ Small companies scale by formalizing what made them great. Then hiring in (like Basecamp did)