AI Search Visibility

For brands being told they need to “optimise for AI” that want to know where they actually stand before paying for monitoring or content. Especially useful if you already rank in search and want to know whether that carries over into generated answers.

Visibility in AI answers is unstable between runs. Ask the same assistant the same question twice and you will often get a different list of brands. In research published by SparkToro and Gumshoe in January 2026, 600 volunteers ran 12 prompts through ChatGPT, Claude and Google’s AI a combined 2,961 times; the chance of getting the same list of brands in any two responses was under 1 in 100. Anyone promising guaranteed placement in AI answers is selling something they cannot deliver.

What can be measured is how often you appear across many runs. We run your commercial queries repeatedly across ChatGPT, Claude, Gemini and Perplexity and report your mention rate and your competitors’, each with its spread across runs, so you can see which differences are real and which are noise.

We also record which sources those answers cite, and compare your presence in generated answers with where you rank in ordinary search. The gap between the two is usually where the useful fixes are: pages that rank but are never cited, or third-party sources the models lean on where you are absent.

You get specific content and technical fixes, and the method itself: queries, run counts and scoring, documented so your team can re-run the baseline later and compare like with like.

What's included

• Commercial queries run repeatedly across ChatGPT, Claude, Gemini and Perplexity
• Brand and competitor mention rates, each reported with its variance across runs
• The sources those answers cite
• The gap between ranking in search and appearing in generated answers
• Specific content and technical fixes
• A documented method so you can re-run the baseline yourself

What this does not include

• Guaranteed placement in AI answers
• Ongoing monitoring
• Content production

How you know it worked

A baseline you can re-run yourself, with the queries, run counts and scoring documented so your results compare directly with ours.

What we need to start

A list of the commercial queries that matter to you, or we propose them for your approval.