The average AI-visibility score for a well-known SaaS brand: about 42/100

When we ran our State of AI Visibility study — 18 recognizable SaaS brands across CRM, project management, email marketing, and help desk, scanned with the real unprompted buyer questions their customers actually type — the average score came out to about 42 out of 100. This post is one idea, isolated: what that single number is actually made of, and why it should worry you even if your own brand is nowhere near the bottom of the list.

What 42 is not

42 is not "named in 42% of the questions we asked." That's the single most common misreading of an average AI-visibility score, so it's worth killing first. The score is a weighted blend — 60% mention rate, 25% position in the shortlist when ranked, 15% citation of the actual domain — computed only over unprompted questions, the ones that don't already contain the brand's name. A brand that's named in most answers but never ranked near the top and never cited directly can still land in the low-to-mid 40s. So 42 isn't "invisible half the time." It's closer to "present, but thin" — mentioned without being the confident, top-of-list recommendation, and rarely handed to the buyer as an actual link to click.

Why the average understates the real spread

An average of 42 across 18 brands hides something an average always hides: the spread. Some of those household names scored dramatically lower than 42 — genuinely close to zero, meaning zero unprompted mentions survived our honesty filters on any of the four engines we tested. Others scored meaningfully higher, holding a real top-three position and a citation on at least one engine. The average is a useful headline number, but if you're benchmarking your own brand against it, the honest comparison isn't "am I above or below 42" — it's "where do I actually sit in that spread, and who's sitting where I want to be." The full leaderboard shows exactly that, brand by brand, industry by industry, with a live "most invisible name on the list" callout that updates as we re-scan.

Why household names still land in the 40s

The instinctive assumption is that big, well-known brands should score close to 100 — they have huge marketing budgets, deep content libraries, and years of SEO investment. They still don't, and the mechanical reason is the same one we've written about generally: AI engines aren't scoring brand recognition, they're scoring evidence found for a specific buyer question. A brand can be famous and still not be the answer an engine reaches for when a buyer asks "best CRM for a five-person sales team that hates data entry" — a narrower, more situational question than "what is Salesforce." Recognition and recommendation are different things, and the gap between them is exactly what a 42-average study makes visible at scale, across brands nobody would call unknown.

The number that should actually worry you

If a curated list of recognizable SaaS names — the kind of brand most founders would assume is "obviously fine" on AI visibility — averages 42, the honest implication isn't about those 18 brands. It's about the base rate for everyone else. If you haven't checked your own score, the safe assumption isn't "we're probably above average because we take marketing seriously." Plenty of the brands in that study take marketing seriously too, and they averaged 42.

See where you'd land in the same study

The methodology behind this number is the same pipeline behind every free scan on this site: 25 real buyer questions, put to ChatGPT, Claude, Gemini, and Perplexity, scored with the identical formula, identical honesty filters, no hand-tuning. Run it on your own domain — no signup, about a minute — and you'll get your real number, not an estimate of where a brand "like yours" might sit. Then compare it honestly against the 42 above, and against the live leaderboard if you want the brand-by-brand detail behind the average.