14
13 Comments

I built an AI copilot for SEO

I spent a year building an SEO tool because all of the current solutions produce AI slop and call that SEO and it's getting on my nerves.

I've been doing SEO consulting for 15+ years. Wrote a book on it. Built an agency. The whole thing.

And honestly? The last few years have felt gross.

Every AI SEO tool that's come out is basically the same thing: "Generate 100 articles a month! Rank for everything!" Meanwhile the articles lower the trust and quality signals associated with the site, and the articles don't seem to rank that well either.

I kept thinking - this isn't SEO anymore. It's just spam.

The thing that actually bugged me

What I actually do as a consultant isn't write articles. It's look at data, interpret what's happening on the client's site, implement best practices, and figure out how to grow the client's traffic (and rankings). That's the entire point.

But there's no AI tool that does that. There's no AI copilot just for SEO.

Developers have Cursor and Copilot and Claude Code - AI that actually understands their work and helps them think. SEO has... article spinners (under the guise of AI).

So I built the thing I wanted

It took about a year. It connects to live search data (keywords, backlinks, SERPs) and you just... talk to it. Ask questions about your site or pages you want to rank. And get actual recommendations (based on best practices)

No content generation. No "optimize this article" buttons. Just: here's what's happening with your site, here's what I'd do, here's how to improve it.

Anyway - it's live now. Curious if anyone here finds it useful or if I've been building in a vacuum.

You can check it out here:
SEO Blab

Happy to answer questions about the tool or SEO in general.

on December 8, 2025
  1. 1

    I’ve felt that pain of tools spitting out bland drafts, so your approach hits home. If you ever expand into location-focused workflows, checking how services like https://crowdo.net/local-seo handle audits, on-page tuning, and local signals might spark ideas for features your copilot could automate or simplify. Local SEO has tons of repetitive steps that an AI assistant could make way less of a grind.

  2. 1

    This resonates so much. The SEO space has turned into a race to generate as much AI sludge as possible, and it’s wild how many tools proudly ship features that actually hurt the sites they’re supposed to help. The fact that you stepped back and said, ‘No—SEO is analysis, not mass content production,’ is exactly the kind of correction the industry needs.
    What you built sounds closer to an actual SEO copilot: something that interprets data, understands intent, and helps people think instead of just dumping filler text. Honestly, that’s the missing layer right now. Developers have smart copilots, marketers have analytics dashboards… but SEO pros? Until now it’s been dashboards + article spinners.
    Really curious to see where this goes. If your tool actually helps people understand their site and make high-leverage decisions, I could see a lot of SEOs switching over. This feels like it was built by someone who’s been in the trenches—not someone chasing the latest buzzword.

  3. 1

    My first impression was that it was for SEO only, but I see that you're helping with GEO as well, which is really interesting today. How do you track mentions across all different LLM models?

    1. 1

      The mentions are checked on both chatGPT and Google AI overviews live. Actually, I've built several AI SEO tools into this tool, because that's becoming such an important part of modern SEO. Such as, it can also see what pages are often being cited/mentioned for a particular topic, so you can see what types of content AI likes to recommend - then go and create that content yourself.

  4. 1

    This comment was deleted 2 months ago.

Trending on Indie Hackers
I'm a lawyer who launched an AI contract tool on Product Hunt today — here's what building it as a non-technical founder actually felt like User Avatar 140 comments “This contract looked normal - but could cost millions” User Avatar 54 comments 👉 The most expensive contract mistakes don’t feel risky User Avatar 41 comments The indie maker's dilemma: 2 months in, 700 downloads, and I'm stuck User Avatar 39 comments I spent weeks building a food decision tool instead of something useful User Avatar 28 comments A simple way to keep AI automations from making bad decisions User Avatar 26 comments