The Contribution Surface

A designer reached out a few years ago. He'd seen the work, liked what we were doing, and offered to help with a t-shirt site we'd started called Hope and Courage. He had the skill. He had the time. He wanted in.

I couldn't give him what he needed to start.

The brand guidelines lived in my head. The site structure was half-documented across three tools. The voice existed as a vibe, not a doc. The process to ship a shirt, what went where, who approved it, what the marketing around it looked like, existed, just not anywhere he could find it.

He wanted to contribute. I'd built a company he couldn't contribute to.

Every founder I know has heard "I'd love to help, how can I help?" and fumbled the answer. We treat it as a social script. Polite exchange, polite deflection, back to work.

The honest version is a product problem. The talented person showed up. The door was unlocked. Inside, there was no map. They drifted. We went back to building, told ourselves we were focused, and the pattern repeated.

The bottleneck isn't willing helpers. It isn't talent in the market. It's the absence of the structure that lets willing, talented people do useful work without you in the loop.

Most of us treat that structure as overhead. Documentation is for later. Brand guidelines are a distraction from shipping. The decision tree lives in my head because it's faster there. All of that is true, right up until someone shows up wanting to help. Then it's the thing that decides whether they can.

Open source software figured this out thirty years ago.

Open any healthy open source project and you can tell within two minutes whether you could contribute. A README explains what the project is. A CONTRIBUTING doc explains how to work on it. Issues labeled "good first issue" or "help wanted" tell you where to start. A test suite tells you whether your work is correct before anyone else has to look at it.

None of that is the product. All of it is the infrastructure that makes the product multiplayer.

The projects that attract the most contribution aren't the ones with the most talented users. They're the ones where the distance between interested and contributing is the shortest.

A lot of founders have adopted "build in public" and stopped there. Post the work. Show the progress. Grow an audience.

Audience is one-way. Contribution is two-way. Posting what you're doing creates followers. It doesn't create collaborators unless you've also built the surface where collaboration can land. The gap between "people like what you're doing" and "people can help you do it" is enormous, and most of us live on the wrong side of it without noticing.

Inside a company, the same pattern as an open-source repo translates directly:

• A single source of truth for what you're building and why (the README)

• A short guide to how new people plug in (the CONTRIBUTING doc)

• A visible list of work someone could take on without asking first (the issues)

• A clear standard for what "done" looks like (the tests)

The format is negotiable. The point isn't. A motivated person arrives, finds their footing, and ships something useful without a thirty-minute onboarding call. That's the contribution surface. Without it, every offer to help is a load on you instead of leverage for you.

The same gap is everywhere once you see it. Nonprofits with communities that would contribute if there were a path. Companies with customers who'd improve the product if improving it were possible. Teams with junior people who could take on more if more were legible.

We tell ourselves the bottleneck is resources. The bottleneck is usually structure.

The designer who offered to help with Hope and Courage eventually drifted. Not because he stopped caring. Because I never built the thing that would have let him in.

The work most of us avoid isn't the product. It's the surface that lets other people help build the product. It looks like overhead. It's the thing that decides whether you're a bottleneck or a multiplier.

Every offer since, I think about what would have needed to exist to say yes.

P.S. I'm testing this on myself. jordanbartlett.io is going live as an open-source project later this week. README, contribution guide, open issues, the whole surface. If the argument in this issue is right, the proof will be whether anyone shows up.

Receipts in the next issue, either way.