We Built an AI-Visibility Tool and Scored 0/100 on Ourselves

We build a tool that measures whether AI assistants recommend your brand when buyers ask. The first thing our founder did with it was run our own products through it.

Every one of them scored 0 out of 100.

Forge: 0. Faultline: 0. Dx3: 0. NextGen AI: 0. When someone asks ChatGPT for the best tool in any category we compete in, the assistant names other companies — and not one of ours.

We could have quietly fixed that before telling anyone. We're going to do the opposite. Here's the scoreboard, and here's us fixing it in public, one weekly re-scan at a time.

What "0/100" actually means

The tool does one thing, honestly. It asks ChatGPT — with web search on — the real buying-intent questions for a category. "Best tool to take an AI prototype to production." "Universal intelligence platform for enterprise data." Then it measures three things: are you cited, how prominently, and who the assistant recommends instead.

There's no keyword theater, no vanity metric. A 100 means you're the answer. A 0 means you're absent — the conversation happens, a decision gets shaped, and your name never comes up.

When we ran Forge, the assistant confidently recommended Wikipedia, TechRadar, and a couple of generic task managers. Not because those are better at what Forge does. Because the model has never been given a reason to know what Forge is.

The uncomfortable insight in our own data

Here's the part that stung, and the part that's most useful.

Look at who the AI recommended instead of us. For Forge — an AI delivery and agent-orchestration platform — it cited task-list and design tools. For Dx3 — an intelligence platform — it cited customer-support chatbots. The model didn't rank us low in our category. It put us in the wrong category entirely.

That reframes the whole problem. You can't win the question "best AI delivery platform" if the assistant doesn't believe you're an AI delivery platform. Step one isn't out-ranking a competitor. Step one is teaching the model what space you're in — through clear category content, structured data it can read, and citations from the sources it already trusts.

This is exactly the pain our agency customers feel on behalf of their clients. We just felt it on ourselves first. That's the most honest product validation there is: the founder saw our own zero and said "fix it," before any customer asked.

What we're doing about it — in the open

No magic, no overnight jump. AI visibility moves over weeks, because it rides content getting indexed and third-party sources citing you. The plan:

1. Define our categories clearly — so the model stops mis-filing us.
2. Publish honest comparison and FAQ content with structured data the assistant can cite directly.
3. Earn placement in the roundups AI already trusts — the same listicles and directories it's citing instead of us.
4. Re-scan every week and publish the delta — including the weeks it doesn't move.

If it works, you'll watch four zeros climb. If a tactic flops, we'll show you that too. That's the point of doing it in public: the scoreboard can't be spun.

You can run the exact same scan

The tool that gave us four zeros is the same one you can point at your own brand. Enter a domain and a category, and you'll see your score, the competitors the AI recommends instead, and where you go invisible.

It might tell you you're the answer. It might tell you you're a zero, like we were. Either way, you'll know the number — which is more than most brands competing in AI search can say right now.

Run the same free scan on your brand →

Free, no signup — takes under a minute.

This is week zero of GEO Score Watch. We'll publish the NXTG portfolio's AI-visibility scoreboard every week as we work to move it — wins, plateaus, and all.