Why I put my company inside ChatGPT before my competitors woke up
For most of the last twenty years, the way a customer found a business like mine was simple. They opened a search engine, typed in what they needed, and picked from whoever had paid or ranked their way to the top. A whole industry grew up around winning that moment.
That moment is moving. More and more people now ask an AI assistant before they ever touch a search box. They do not want ten blue links and four quotes to compare. They want an answer. And the businesses that show up inside the assistant, at the point the question gets asked, are going to own the next decade of customer acquisition the way the SEO winners owned the last one.
So this week we did something about it at JustFix, the UK home-services company I run.
What we shipped
JustFix is now live as an official app in the ChatGPT app store. Someone can ask ChatGPT how much it costs to fix a dripping tap or change a lock, and get a real, itemised quote from us and a link to book a vetted tradesperson, without ever leaving the conversation.
Underneath it sits a public MCP server. MCP is the open standard that lets AI assistants connect to real-world services, and we built ours so that any MCP-capable assistant, not just ChatGPT, can quote a JustFix job live. We also packaged it as open-source skills for the leading agent frameworks.
I should be honest about why this one was fun to build. I am an avid user of OpenClaw and Hermes, the agent frameworks I run my own day-to-day on, so making my first ever skill, and making it one that those tools can actually use, was a genuine pleasure rather than a chore. There is something satisfying about building for a community you are part of. The skill is open source on GitHub at github.com/Just-Fix/justfix-skill, and the MCP server is listed in the public directories, including mcp.so. If you use agents yourself, you can plug JustFix in today.
Why I am writing about it here
I am not posting this to show off a feature. I am posting it because of what it represents, and because I think every owner reading this should be asking the same question about their own business.
A few reflections.
Distribution is shifting again, and shifts like this reward the early. The owners who win are rarely the ones with the best product. They are the ones who show up in the new channel before it is obvious, while it is still cheap and uncrowded. MCP and the AI app stores feel a lot like search did in 2004, or the App Store in 2009. Early is an advantage you only get once.
Small teams can move on this faster than big ones. JustFix is not a giant. That is exactly why we could decide to build on MCP and ship it in weeks, while larger competitors are still forming a committee to discuss it. If you run a lean business, this is your edge, not your excuse.
It is an exit signal as much as a growth lever. When I look at JustFix the way an acquirer would, being natively present in the AI channel is the kind of thing that makes a business look forward-facing rather than legacy. It is a story a buyer pays a premium for. The work you do to be early is the same work that makes you irresistible later.
The lesson I keep coming back to is the one I write about most: build for where your market is going, not where it has been. The search box is not dead, but it is no longer the only front door. We just made sure JustFix is standing at the new one.
If you want to see it in action, the JustFix app is live in ChatGPT now, and you can try it here.