Four 5-star ratings, two reviews

What a visibility tracker sees when it looks at Axis Suite right now: a 5.0 out of 5 rating on Capterra, on G2, on GetApp, and on Software Advice. Four review sites. Perfect scores across all of them. By the logic most AI-visibility tools use, that reads as strong, well-corroborated proof.

It is two reviews.

Perplexity flagged the thin evidence itself when we asked whether Axis Suite was any good.

What it did not flag is that three of those four sites, Capterra, GetApp, and Software Advice, share one review catalog, so their three listings are one review, not three. Spotting that takes resolving the relationship between the sources, which is the part neither a model nor a mention-counter does on its own.

A review submitted to any one of them appears on all three. As of 2026 they also share a single owner, since G2 acquired all three from Gartner, but the catalog was pooled long before that. So four 5-star listings are really one review shown three times, plus a second review on G2. Two reviews, two independent judgments.

Here is what our own tool does with that, on our own data.

It counts two independent judgments, not four. The Gartner Digital Markets catalog, Capterra plus GetApp plus Software Advice, collapses into one. G2 stands as a second. Then it adds the part the badges hide: two judgments, each resting on a single review, is thin evidence. It scores that 20 out of 100 against a target of ten independent judgments, and it labels the target an assumption, not a law.

That number is smaller than the flattering one. That is the point. We would rather show two honest judgments than four inflated ones, even when all four are ours.

A citation is not evidence. A repeated citation is not repeated evidence. Most tools count the badges. The number that matters is how many independent judgments actually sit underneath them, and for most brands, ours included, that number is smaller than it looks.