The Stewart Butterfield
Framework
Product Taste & Iterative Design
Stewart Butterfield built products and companies by reasoning from first principles in industries that the incumbents considered mature. Stewart Butterfield's product and strategy thinking, captured in this framework as mental models, decision heuristics, and the signature questions they would ask.
The product taste & iterative design framework distils Stewart Butterfield's documented approach into a .md skill file you can load into Claude, ChatGPT, or any LLM in seconds.
Where most AI personas generate generic output, this framework is grounded in real documented work: interviews, writing, published frameworks, and the specific decisions Stewart Butterfield made when it mattered. Use it for decision-making, writing, strategy, or any situation where you want your AI to think with Stewart Butterfield's distinctive patterns rather than a default model voice.
A glimpse of Stewart Butterfield's thinking.
Great products emerge from the tension between what users think they want and what they actually need once they experience it. The best product decisions come from obsessive attention to how real people behave in your product, not from roadmaps or competitor analysis.
- 01
- 02
- 03
- 04
- 05
Five specific models, each named and explained with concrete examples, unlocked with purchase.
"What are our most engaged users doing that we didn't expect?"
4 more questions in Stewart Butterfield's voice are unlocked with purchase.
Full list of anti-patterns unlocked with purchase.
The full framework adds: detailed explanations of all five mental models with real-world examples, the decision heuristics Stewart Butterfield actually used, remaining signature questions, the complete list of anti-patterns, a copy-paste activation prompt, and a worked example.
Unlocked the moment you purchase. ~2,000 words total. Delivered as a Claude Skill and plain .md within 60 seconds of checkout.
- ✦Core philosophy: 2–3 sentences grounded in Stewart Butterfield's documented worldview
- ✦Five signature mental models: each named, with concrete examples from their work
- ✦Decision heuristics: the concrete rules they actually used
- ✦Signature questions: five questions that sound like Stewart Butterfield, not a generic MBA
- ✦What they would NOT do: the anti-patterns this framework rejects
- ✦Copy-paste activation prompt: drop straight into Claude or ChatGPT
- ✦Worked example: the framework applied to a realistic scenario
- →stewart-butterfield.zip: a native Claude Skill. Upload to Settings → Skills in Claude, or unzip to
~/.claude/skills/ - →stewart-butterfield-framework.md: plain markdown. Works in ChatGPT Custom GPTs, Gemini, Claude Projects, agent system prompts, anywhere
What it does with a real question.
A founder is deciding whether to take a large but difficult client.
Take the client, but only on terms that let you learn from them. A large, difficult client is a loud, concentrated source of product truth, and the question is whether you can turn that signal into something your whole market needs or whether it will just turn you into their custom shop.
Here is the test I would apply. Difficult clients usually fall into two groups. Some are difficult because they are early: they hit the edges of your product first, hack around its gaps, and ask for things a hundred later customers will also want. Others are difficult because they are idiosyncratic: their demands trace back to their org chart, their legacy systems, their politics. The first kind is a gift. The second is a tax that looks like revenue.
Think of the workarounds they invent as your next feature. If they are building spreadsheets, exports, and manual processes around a real gap, and you can imagine smaller customers doing the same, lean in. If every request needs a sentence beginning "at our company, we always," walk away.
Next move this week: ask for a paid, ninety-day pilot with a narrow scope. Write down the three biggest requests before you start, then check at day sixty whether any of them came up with your other customers. If none did, you have your answer, and you have only spent a quarter finding it out. Do not rearrange your roadmap around a signature until then.
The full framework carries five models, decision heuristics, signature questions and a worked example.
Written by claude-sonnet-5-5 with this framework loaded, 2026-10-11. Checked automatically for length and for copied framework text.
Save up to 40% with a bundle
Works as a Claude Skill (.zip) and as a plain .md for ChatGPT, Gemini & every LLM
Delivered to your email within 60 seconds · Secured by Stripe
Price includes UK VAT where applicable
Inspired by the documented thinking of Stewart Butterfield. A mental-model toolkit, not a literal representation. Delivered as a .md file compatible with Claude, ChatGPT, Gemini, and any LLM.