Do Things That Don't Scale
Hand-serve your first users obsessively to discover what millions will love
The most counterintuitive rule of scaling is that the first step is to renounce your desire to scale. Before you can serve millions, you must obsessively hand-serve a tiny cohort of users until you understand exactly what makes them love your product. It is better to have one hundred users who love you than a million who just kind of like you, because fanatical users stick around and tell their friends while lukewarm users vanish.
This framework rejects the MBA instinct to 'go big' from day one with marketing blitzes and viral growth hacks. Instead, it demands that founders physically go to their users, meet them one by one, listen deeply, and handcraft the product based on direct feedback. The small, painstaking work of getting every detail right during this phase creates a narrow but deep wedge of loyal users that becomes the foundation for explosive growth.
The 11-Star Experience technique, developed by Airbnb's Brian Chesky, extends this approach by forcing teams to imagine wildly impractical levels of delight (from a welcome gift at 6 stars to going to space at 11 stars) and then working backward to find the sweet spot between ordinary and extraordinary. This design exercise consistently produces breakthrough product ideas that differentiate from competitors.
- It is better to have 100 users who love you than 1 million who just kind of like you
- Fanatical early users are the foundation for compound growth; lukewarm adoption is the illusion of scale
- The roadmap for your product often exists in the minds of the users you are designing for
- Ask users not 'What can I improve?' but 'What would make you tell everyone you know?'
- The handcrafting phase is the most creative period of a company's life and cannot be recovered once passed
- Design the extreme (the 11-star experience) and then work backward to find the practical sweet spot
- Go to your users physicallyTravel to where your users are. Meet them in person or via video call. Do not rely on surveys, analytics dashboards, or support tickets alone. When Airbnb's founders went to New York and literally stayed in hosts' homes, they discovered product insights that no amount of remote data analysis could have revealed.
- Serve them one by oneOnboard users individually. Answer support calls yourself. Do the manual work that you will later automate. Melanie Perkins of Canva personally onboarded early users, walking them through the product step by step. This manual process revealed friction points that automated onboarding would have missed entirely.
- Ask transformative questions, not incremental onesInstead of asking 'What can I do to make this better?' which yields small suggestions, ask 'What would it take for you to tell every single person you've ever met about this?' This reframes the conversation around delight rather than adequacy and surfaces breakthrough ideas.
- Design the 11-star experienceTake one part of your product and map out what a 1-star through 11-star experience looks like. The 1-3 star range covers failures, 5 stars is adequate, 6-7 is delightful, 8-9 is extravagant, and 10-11 is absurd fantasy. The sweet spot between adequate and absurd is where your competitive advantage lives. Work backward from the extreme to find it.
- Identify and empower your superfansWatch for the users who produce binders full of suggestions, who 'forge' workarounds, who use your product in unexpected ways. These superfans are your most valuable product advisors. Give them channels to share feedback and treat their detailed suggestions as your product roadmap.
When Airbnb wanted to launch Trips (curated travel experiences), Brian Chesky's team did not build a platform first. They posted a flyer seeking a single volunteer traveler. Ricardo from London was selected. His self-planned San Francisco trip was a typical tourist experience of lines and loneliness. Airbnb then worked with a Pixar storyboard artist to script a transformative trip for Ricardo, featuring host dinners, restaurant reservations, and a midnight bike tour. By the end, Ricardo was in tears calling it the best trip of his life.
In 2009, Brian Chesky told Y Combinator's Paul Graham about his grand vision for Airbnb. Paul had no interest in spreadsheets or projections. He asked where the users were (New York), noted that Brian was still in Mountain View, and said: 'Go to your users. Get to know them. One by one.' Brian protested that meeting users individually wouldn't scale. Paul replied: 'That's exactly why you should do it now. It's the only time you'll ever be small enough to meet all your customers.' Brian and co-founder Joe went to New York, visited hosts in their homes, photographed listings, and discovered a host with a binder containing dozens of pages of product suggestions. That binder became Airbnb's product roadmap.