Notes on AI search visibility, from building Foreground.
We asked ChatGPT, Gemini, Claude and Grok the same questions for 8 trades in 3 Dutch cities. They named 723 businesses between them. Only 41 were named by all four.
October 13, 2026Everyone wants to be named when someone asks ChatGPT. An AI answer has room for a handful of names. An honest look at what a tool like Foreground can and can't do about that, from the guy who built it.
October 8, 2026Foreground's old 0-100 score gave almost every small business a 0. Before replacing it, we asked 11 customer niches' questions to four AIs, twice, and wrote the pass rules down first. Here's what came back and what we built from it.
September 17, 2026A second self-scan, four days after "we-scanned-ourselves," found the same disambiguation problem — plus a new one. Two models described detailed, confident, entirely fictional products called Foreground. Neither exists.
September 13, 2026A real self-scan turned up a problem no amount of schema markup fixes — most models don't know which "Foreground" we mean, and one partly made up a fake version of our own company.
September 9, 2026A grounded look at the AI-visibility/GEO tool landscape — Profound, Otterly, Peec AI, Scrunch — and where a free, four-model scan like Foreground fits.
September 2, 2026A free, 40-minute checklist for checking and improving how your business shows up in ChatGPT and Gemini answers, no tool required for the first four steps.
August 29, 2026SEO wins a ranking. AI visibility wins a sentence. Why the two are mechanically different, and what actually changes what a model says about you.
August 26, 2026An AI answer isn't a ranking, it's a cloud of candidate businesses compressed into one sentence, and vague businesses lose that compression first.
August 22, 2026AI answers don't give you a ranked list, just one or two names, and good SEO doesn't guarantee your business is one of them.