# If We Publish Our API Documentation Our Competition Will Just Copy What We Do

**Published:** 2026-08-13  
**Author:** Kin Lane  
**Canonical:** https://apievangelist.com/2026/08/13/if-we-publish-our-api-documentation-our-competition-will-copy-us/

I wrote yesterday about how [people tell me they have partner APIs instead of public APIs](https://apievangelist.com/2026/08/12/we-have-partner-apis-not-public-apis/), and continuing on this same theme of excuses I hear from API producers, I wanted to explore the reasons why the documentation for APIs is not available. The most common one I hear from folks is that if they publish their API documentation, their competition will just copy what they offer -- so they make it available on request.

To answer this one I always go back to the example I provided to Google's lawyers in the Oracle vs. Google API copyright case over a decade ago. Your API documentation is the menu for your restaurant. Can you imagine if your favorite burger restaurant said, "we won't publish our menu because our competition will copy us," and made you request it instead of publishing it online, in the window, and just bringing it to you when you sit down? Your API documentation is not your APIs. You aren't giving anything away.

Just like the hamburger, what you offer isn't just what is on the menu. It is the ingredients and the process you use. And just like the hamburger, you can't order anything from the menu if you don't pay for it. Your API documentation is your menu. Having it publicly available, consumable by humans and agents, is how we do business online in 2026. Hiding your menu isn't a business strategy. It just reveals that you probably don't have a strategy that is well articulated, easily communicated, defensible, and unique to your operations. Having your menu published publicly where anyone can find it demonstrates a certain level of confidence in what you do.

If your API documentation isn't public it isn't being indexed by search engines and the waves of crawlers producing training data for all of the leading large language models (LLM). This means that the artificial intelligence tooling that you use, that your competition uses, and that your customers use doesn't natively speak your APIs. Sure, you can always feed this in via MCP and Agent Skills, but why not ensure they speak it natively, and then defend your operations using modern authentication, authorization, and monetization techniques? It shouldn't matter that your competition is copying you. You should be confident in what you do. You should do it better. And you should have the infrastructure in place that determines who has access to what is on the menu, and who does not.