Why I Built This

In today's world of GEO and AI search, most companies are making a big mistake.

They're tracking prompts like “best SEO agency” or “best GEO agency” — and while that has its place, it misses how people actually make decisions, especially in B2B.

A Gartner study found that 77% of B2B buyers ask people in their network for recommendations. That means by the time someone turns to AI, they probably already have names. Their friend said “check out Seer Interactive” — now they're typing “Seer Interactive vs [competitor]” into ChatGPT or Perplexity.

That's a fundamentally different prompt than “best SEO agency.” And it's not how most GEO agencies are tracking.

I thought that was a major gap. So I vibe coded this tool.

How to Use the Output

I know this is probabilistic and rife with issues. The goal is not to build a perfect tracker. The goal is to find the big gaps.

When you run a comparison, look at the themes where a competitor might outperform you. For instance, if we at Seer consistently score low on “speed” against competitors across multiple models — and I can see that pattern — then maybe that's a signal to start building content around our speed of execution.

The output shows you how AI perceives your brand head-to-head against competitors, given the kinds of prompts I believe real people are actually entering. Use it to find the gaps worth closing, not to obsess over individual scores.

Help Me Make This Better

This is my first ever public GitHub repo. I'm putting it out there because I think other marketers and agencies need a tool like this, and I'd rather build in the open than keep it to myself.

If you have ideas, find bugs, or want to contribute — I'd genuinely appreciate it.

Built by Wil Reynolds at Seer Interactive