Almost every company shows up somewhere. That is not the same as showing up where it counts.

During the study window, from 2026-07-01 to 2026-07-31, 96.1% of the 153,251 companies in the cohort were named by at least one of the three engines, ChatGPT, Claude, or Gemini, in at least one answer. Put the other way: 0 of 153,251 companies never appeared on any engine during the entire window.

This is the single most reassuring number in the study, and it is also the one most likely to be misused. It is reassuring because it rules out the fear that AI assistants have simply never heard of most businesses; whatever material these engines draw on when assembling an answer, it is broad enough to include almost the entire measured cohort at least once. It is easy to misuse because "named at least once, by at least one engine, somewhere in the answer" is a very low bar, and clearing it says nothing about whether a company is named consistently, named early in the list, or named by the specific engine its actual buyers happen to use.

Why the bar is low. The measurement behind this figure counts a single mention, in a single position, on a single engine, across the full month, as sufficient to count a company as visible. A company that appeared once, in eighth position, on one engine, on one day of the month, counts exactly the same in this figure as a company that appeared first, on all three engines, every time it was asked. The two companies have very different commercial outcomes from that visibility, but this particular statistic cannot tell them apart. That distinction is what the average position figure and the cross-engine agreement figures, covered elsewhere in this study, are built to capture.

What would make this number lower. If engines were systematically excluding certain categories or business types, this number would show it, because the denominator is the full cohort and the exclusion would show up as companies stuck permanently in the 0 figure. It does not show that pattern this month: the count of companies never appearing anywhere is 0, out of 153,251. That is consistent with, though not proof of, a retrieval process broad enough to surface nearly any company that has a reasonable public presence, rather than a process that pre-filters by some hidden quality judgment before a company is even eligible to be named.

What a company should actually take from this. If a company suspects it has been excluded from AI search entirely, this figure says that outcome is uncommon in this cohort, so the more useful next question is not "am I visible at all" but "am I visible on the engine my buyers use, in a position they would actually notice." Those are the questions the rest of this study addresses, and they require the position and cross-engine data, not this topline figure, to answer.