ProBackend
News & Analysis
4 days ago6 min read

What Semrush’s AI Visibility Awards Can—and Can’t—Tell You

What Semrush’s AI Visibility Awards Can—and Can’t—Tell You

Semrush’s AI Visibility Index separates brand mentions from citations. Here’s how to interpret its awards and what visibility does—and doesn’t—prove.

What the Semrush AI Visibility Awards measure

Semrush’s AI Visibility Index presents its awards in the context of a large-scale benchmark of brand presence in AI answers. Its index says it analyzed more than 126 million real US AI-search prompts across 22 industries and four platforms, and identifies a set of “Universal 36” brands that won every month in its study period. That is the core finding to take from the awards: a group of brands repeatedly achieved strong visibility across the benchmark, rather than a one-off mention being treated as proof of lasting leadership. Source: Semrush AI Visibility Index

The index defines a brand mention as the appearance of a brand name in an AI-generated answer. Its leaderboard ranks the top 20 most-mentioned brands and can be filtered by date, industry and AI platform. These measures describe visibility within the prompts and answers Semrush observed. They do not, by themselves, establish that users trusted a brand, chose it, or purchased from it.

The distinction matters because “winning” can sound like a universal verdict. The awards are better understood as recognition within a particular measurement framework: a defined prompt sample, platforms, sectors and time window. The recurring Universal 36 finding is notable precisely because persistence is different from a single high-ranking snapshot. But it is still a statement about observed AI answer presence, not a ranking of overall brand quality.

Mentions are not the same as citations

Semrush explicitly separates mentions from citations. Mentions count how often a brand name appears in an AI answer. Citations count how often the brand’s domain is cited as a source. The index describes the first as a visibility measure and the second as an authority signal. A brand can therefore be talked about without its own site being cited; conversely, a site may be cited frequently without its brand dominating the answer. Source: Semrush AI Visibility Index

That separation is essential when interpreting award results. A high mention count can indicate that a model includes a brand in the conversation, but it does not tell us why the model named it or whether the answer endorses it. A citation provides evidence that a domain was used as a source in the answer, but it is not a direct measure of accuracy, trustworthiness or user confidence. Those are stronger conclusions than the underlying counts warrant.

The index also shows brand performance in relation to authority as a cited source. Its explanation describes four possible patterns: high visibility with low citation authority means the brand is discussed without being cited; high authority with lower visibility means it is a source without dominating the conversation. This makes the awards more useful as a starting point for analysis than as a single score to celebrate. Ask which kind of presence the recognition reflects, and whether that presence fits the brand’s goals.

What the benchmark can reveal

The index’s scope offers several useful lenses beyond a headline list. It reports brand mentions, competitor co-occurrence, sources cited by AI, and comparisons across ChatGPT, Google AI Mode, Google AI Overviews and Gemini. Results can be examined by industry and platform, while the source-analysis feature ranks domains cited in answers. Semrush also compares overlap among leading brands and cited sources across platforms. Source: Semrush AI Visibility Index

These views help explain why an award should not be read as a complete account of AI search. A cross-platform leader may not lead in every individual product or vertical. Semrush’s platform analysis compares the top 100 mentioned brands and top 100 cited sources across platform pairs; overlap indicates similarity in the lists, not that the systems produce identical answers. The index says that comparing these patterns can reveal where platforms rely on different brands and sources.

Co-occurrence adds another useful dimension: Semrush defines it as how often brand pairs appear together in AI answers. That can show which competitors share the same conversational territory. It does not, on its own, indicate whether the relationship is favorable, whether one brand is preferred, or what caused the pairing. The answer text and the context of the prompt still matter.

Likewise, a citation-source leaderboard can point to domains that AI answers draw on, but it should not be confused with a list of brands that won an award. Semrush defines a source as a website domain used to formulate an answer, typically shown as a citation or link. Source counts represent citations across the prompts in the study. They offer evidence about the observed source landscape, not proof that a specific editorial or marketing action caused a brand’s visibility.

How to use an award without overstating it

Treat the award as a benchmark signal, then inspect the dimensions behind it. First, confirm the period, platform and industry represented. The Semrush index describes filters for date, vertical and AI platform, and its published study overview refers to four platforms and 22 industries. An aggregate result can conceal meaningful differences among those slices. Source: Semrush AI Visibility Index

Second, separate mention performance from citation performance. If the objective is to understand whether a brand appears in AI responses, mention counts are directly relevant. If the question is whether its domain is being used as supporting material, examine citations. Neither metric alone demonstrates trust or commercial impact, and they should not be substituted for each other.

Third, read the surrounding answer and source context before drawing conclusions. A brand’s name may appear in a comparison, a caveat or a negative context. A source citation records that a domain was referenced; the count alone does not explain the role it played. The index’s own distinction between visibility and authority is a reason to examine both the numbers and the underlying examples.

Finally, use repeated observations rather than treating one snapshot as a trend. Semrush presents monthly updates and highlights brands that appeared as winners every month in its study period. That recurrence strengthens the case that those brands were consistently visible within this particular dataset. It does not make the result permanent: prompts, platforms, answer behavior and the period being measured can change. Record the study window and compare like with like when tracking movement.

What the awards do not prove

The available index describes a measurement of AI-generated answers, not a randomized experiment or a measure of downstream user behavior. It does not establish that an award caused traffic, conversions, revenue or trust. It also does not prove why a brand was mentioned, or that a particular optimization tactic produced the result. Those conclusions would require evidence beyond a leaderboard or aggregate count.

Semrush’s own framing is most useful when taken literally: mentions measure visibility, citations measure how often a brand’s domain is cited, and the two can diverge. The AI Visibility Awards can identify brands that stood out in the measured environment and prompt sharper questions about platform and source patterns. They are not a substitute for evidence about how people interpret AI answers or what they do afterward.

For marketers, the practical takeaway is to treat the recognition as a map of observed presence, not a certificate of universal trust. Investigate where a brand is named, where its site is cited, which competitors appear alongside it, and how those patterns vary across platforms and industries. Keep conclusions tied to the study’s scope. That approach preserves what the awards genuinely offer—a comparative signal of AI-answer visibility—without turning it into a claim the metrics cannot support.

the semrush ai visibility awards measurethe semrush ai visibility awards measure