“So who’s actually going to own the AI-search stuff around here?”
Ever watch a meeting go quiet the second that question lands?
Do you bolt it onto the SEO person’s plate, or stand up a dedicated AI person? Plenty of teams never settle it, and the whole thing slides to “let’s pick it up next time.”
I keep hearing the same story near me too: the “AI project” gets its own flag planted, and then goes months without a single conversation with the existing search team.
But an analysis of 126 million US AI-search prompts found that how you answer “whose job is this” splits your results in a way that’s hard to ignore.
So let’s take this all the way to the practical bit — how to actually set up the team.
The Study That Peeked At 126 Million AI Searches
What we’re looking at is the expanded “2026 AI Visibility Index” that Semrush published on June 26 2026 (the release).
The scale is real: 126 million AI-search prompts made in the US, analyzed from January through April 2026.
It covered four surfaces — ChatGPT, Gemini, Google AI Mode, and Google AI Overviews — and built benchmarks across 22 industries.
One thing to keep in mind: Semrush sells AI-visibility tracking tools. So this is research by a company with a commercial stake in the conclusion “you should be measuring AI visibility.”
Take the raw numbers at face value, but discount where they lean on the gas.
Merged Teams: 81%. Separate Teams: 36%.
The part that hits the “whose job is this” question head-on is the study’s section on how organizations run the work. It splits organizations by how they handle SEO and AI visibility, then compares how each group reports on results.
- Organizations that run SEO and AI visibility together: 81% report a rise in traffic or leads
- Organizations that manage them separately: only 36% do
That’s a 45-point gap. Same “we’re doing AI stuff,” and the placement alone opens up a canyon.
Here’s the caveat. These 81% and 36% figures come from a survey, so nobody proved that merging causes the lift. It could just as easily be that the organizations already firing on all cylinders happen to keep both under one roof.
Still, 45 points isn’t a gap you wave away. And why splitting them stalls out shows up in the study’s other numbers.
AI Visibility Isn’t “Show Up Once And You’re Done”
The first reason splitting stalls: AI visibility isn’t a one-and-done job.
The study tracked 1,200 brands, and only 36 of them held a monthly top-100 spot across all four platforms for the whole period.
The lineup: YouTube, Google, Reddit, Amazon, Facebook, Apple, Walmart, Disney, Nintendo (nine major-leaguers wandering into a pickup game).
So even if you run a one-off push once a quarter and get your name to surface, plan on it having slipped by the next month.
Work that has to run continuously belongs in the team that already runs things continuously. A newly stood-up experiment usually isn’t built to keep going.
ChatGPT Cites 15 Sources. Gemini Cites 3.
The second reason: each AI cites in a totally different way.
- ChatGPT: about 15 sources cited per answer
- Gemini: about 3 sources per answer
A 5x spread. And it’s not only the count that shifts per engine.
“Mention” is whether your name shows up in the answer; “citation” is whether you get linked as a source. On Gemini, the overlap between those two can drop to around 30%.
So even if you check ChatGPT, see yourself, and conclude “we’re visible,” Gemini can be telling a different story. If you’re going to watch all four continuously, adding a column to your existing search report is faster than anything else.
Measure Your Category’s “Going Rate” Before You Set A Target
The third reason: to set the right target level, you first need to know your category’s going rate. The study lines up 22 industries, and how much of a category’s visibility the top three brands eat up varies wildly by industry.
| Industry | Top-3 brands’ share of visibility |
|---|---|
| News / media | 82.9% |
| Consumer electronics | 76.9% |
| Industrial | 42.2% |
| Finance | 41.4% |
In news and media, the top three take 82.9% of the category. Trying to break in from fourth place there is a serious climb.
Finance sits at 41.4%. More than half is scattered outside the top three, so there’s plenty of room to squeeze in. Same goal of “get more AI exposure,” and the fitness it demands is completely different.
And who’s holding that sense of the going rate? Usually the existing search and content team. Carve AI visibility out into a brand-new department and you leave all of that behind.
Conclusion: Fold AI Visibility Into The Same Search Report
What you carry home from 126 million prompts is that AI visibility doesn’t look like a “stand up a new specialist team” job. Merged operations at 81% against split operations at 36%. That it’s a game of continuity, and that each AI cites differently, both sit badly with a one-off experiment.
So there are two moves. Fold your AI-visibility metrics into the search and content report you already send, as one more column. Then, before you set a target, measure how concentrated your category is. The level you should aim for is genuinely different in an industry where the top three own 80% and one where they own 40%.
One last caveat. This study covers US prompts only, hasn’t been checked in Japanese, and its metrics are Semrush’s own — so they don’t line up directly against another tool’s numbers. Even so, the finding that team placement splits results is worth knowing before you go build a new box for it. Might be worth a look.
Sources
- Semrush, “Semrush Releases Expanded 2026 AI Visibility Index, Analyzing 126 Million AI Search Prompts”, June 26 2026, semrush.com (126 million US AI-search prompts, January–April 2026, across four platforms — ChatGPT, Gemini, Google AI Mode, Google AI Overviews — with 22-industry benchmarks. Organizations running SEO and AI visibility together reported a traffic/lead rise 81% of the time versus 36% when split. Only 36 of 1,200 brands held a monthly top-100 spot across all four platforms. ChatGPT cited about 15 sources per answer versus Gemini’s 3. Top-3 brands’ visibility share: news/media 82.9%, consumer electronics 76.9%, industrial 42.2%, finance 41.4%. Semrush is a vendor of AI-visibility tracking tools with a commercial interest in the conclusion; the 81%/36% figures are self-reported survey correlations.)
- Semrush AI Visibility Index (the index itself), ai-visibility-index.semrush.com
