Find out what AI assistants tell buyers about you.
Searchscope asks the questions your buyers ask across six AI assistants, every week, and keeps every answer word for word. You see where you were recommended, where a competitor was recommended instead, and which sources each assistant used to reach that answer.
-
ChatGPT
named you in 18 of 48 answers, cited you in 8 38% -
Perplexity
named you in 22 of 48 answers, cited you in 16 46% -
Claude
named you in 10 of 48 answers, cited you in 2 21% -
Google AI Overviews
no mention on 24 answers this run 0% -
Gemini
named you in 14 of 48 answers, cited you in 4 29%
Mention rate, this run5 of 6
Six assistants, the same questions, every week.
A one-off check tells you where you stood on a single day. Answers move as assistants update, as competitors publish and as the sources those assistants lean on change. Searchscope runs the same set of questions on a weekly schedule, so you’re reading a trend rather than a snapshot.
-
ChatGPT
-
Claude
-
Gemini
-
Perplexity
-
Google AI Overviews
- Google AI Mode
- The same questions to each oneEvery assistant gets an identical set of prompts, so a difference you see is the assistants disagreeing about you rather than the questions changing between them.
- Asked twice, where the answer can moveChatGPT, Claude, Gemini and Perplexity answer the same question differently on different days, so each question goes to them twice per run and both answers are kept. Google AI Overviews and AI Mode are read once, because they return a result page rather than a fresh answer.
- History survives the list changingWhen an assistant is added, it starts appearing in your runs. When one is retired, the runs it took part in stay in your history, so a trend line never loses its past because a list changed.
Asking about you isn’t the same as asking about the problem you solve.
Those two questions measure different things, so Searchscope keeps them in separate libraries and reports them separately.
- Brand questions Questions that name you. These measure recall: when somebody asks an assistant about you directly, does it know who you are and describe you accurately? Useful, but it’s the easiest test to pass.
- Non-brand questions Questions that name nobody, of the “best tool for” kind your buyers actually type. This is where you find out whether you get recommended when nothing prompts the assistant to think of you.
- Product questions Questions about one product line, tracked as their own library so a strong line doesn’t hide a weak one behind an average.
Group your questions the way you sell.
Put your questions into groups of your own naming, by product line, service or market, and assign them in bulk rather than one at a time. Every view respects the grouping, so you can read the overview, the answers and the citations for one part of the business on its own.
- An average hides the thing you needed to seeA brand doing well across most of its range can be losing badly in one line. Grouped, that shows up as its own row instead of disappearing into a single number.
- Put a bell on the groups that matterTurn alerts on for a group and Searchscope emails you when it drops sharply between runs, so you don’t have to keep checking the ones that are steady.
- Small groups don’t trigger false alarmsA group needs at least three questions asked in both runs before it can raise an alert, so a couple of questions moving can’t page you.
Suggested questions arrive with their reason attached.
Searchscope proposes questions to track and labels each one with where it came from, so you can judge whether it belongs in your set before you add it.
- Already cited you
- You rank for it
- Competitor gap
- Generated
The answers, word for word.
Searchscope keeps the full text of every answer, not a summary of it. For each one you can see which brands were named, how early yours appeared, and whether the assistant linked to you or to somebody else.
“For a small agency, most people land on one of three. Competitor A is the usual recommendation for its reporting, Competitor B is cheaper per project, and your brand is worth a look if you want the analysis rather than just the numbers…”
- NamedCompetitor A, Competitor B, you
- Your first mentionthird of three
- Linked to youno
- Sources usedg2.com, reddit.com, competitor-a.com
Who gets cited instead of you.
Every domain an assistant leaned on to answer is collected across your runs, so you can see which sources shape the answers in your category. Domains that already cite you are listed apart from the ones that don’t yet, because those are two different jobs.
- How much it coversThe number of your tracked questions where that domain appeared in the answer, so a source that shapes twelve answers is easy to tell from one that shaped a single answer once.
- Which assistants use itA domain leaned on by four assistants is a different opportunity from one that only Perplexity reaches for.
- Whether it already mentions youEach domain is marked according to whether it currently cites you, cites a competitor you track, or neither.
- How long it’s been thereWhen the domain first appeared and how many runs it’s shown up in, which separates a durable source from a one-week result.
- The pages doing the workThe specific URLs on that domain that assistants cited, rather than the domain on its own.
- Name variants, caughtWhen an answer refers to you by a spelling or variant you haven’t registered, it’s raised for you to accept or dismiss, so those mentions aren’t quietly scored as misses.
One score, and the five things inside it.
Most AI visibility scores don’t say what they measure. This one does, and the weights don’t change between your dashboard, your reports and your history.
| Component | Weight | What it measures |
|---|---|---|
| Share of voice | 35% | Your share of the brand mentions across answers, measured against the competitors you track. |
| Citation rate | 20% | How often an assistant links to you in the answer, not just names you. |
| Mention rate | 15% | How often you’re named at all. |
| Breadth | 10% | How many of the assistants mention you, so strength on one doesn’t read as strength everywhere. |
| Branded recall | 20% | On questions that name you, how often the assistant knows who you are. |
- Branded recall is measured on its ownAnswers to questions that name you are scored separately as recall, so you can see how much of your standing comes from people who were already looking for you.
- Not asking never reads as not being knownIf you track no brand questions, that 20% is shared out across the other four components rather than counted as a zero, so the score reflects what you actually measured.
Your brand against the brands you’re measured against.
AI answers are one signal. Alongside them, Searchscope benchmarks you and each tracked competitor on the things that tend to feed those answers in the first place.
- Mentions in AI answers, run by runEvery run records how many answers named you, how many cited you, and the same for each competitor, so you can watch a gap widen or close instead of comparing two screenshots.
- Links and authorityReferring domains, total backlinks and domain rank for every brand in the set, side by side.
- Media coverageTotal mentions and news mentions, the domains carrying them, and the split between positive, negative and neutral coverage.
- Branded search demandHow many people search for each brand by name, month by month, which is the closest thing to a direct read on brand awareness.
See a competitor’s PR campaign while it’s still running.
New and lost linking domains are tracked month by month for every brand you benchmark. When a month runs well above that brand’s normal rate, it’s flagged as a burst.
- A burst says somebody ran somethingThe monthly view shows you when. Investigating the burst shows you what.
- The campaign asset, identifiedSearchscope pulls the new links from that window and groups them by the page they point at, which is usually the thing the campaign was built around.
- Coverage separated from plumbingFooter links, legal pages and naked-URL anchors are counted apart from earned coverage, so the number you read is the coverage rather than the boilerplate.
- A counter angle to considerEach investigation is labeled with the kind of campaign it looks like, alongside a suggested angle of your own.
A second advisor, pointed at your brand rather than your rankings.
The PR Strategist reads your AI visibility runs, your brand benchmark, your PR activity and your citation gaps together, then works through what to do about them with you.
- Why the answers favor somebody elseAsk why assistants recommend a competitor over you and get an answer grounded in your own runs, rather than a general theory about how AI search works.
- Who’s worth approachingTurn the citation gaps into a ranked list of sources: the domains shaping answers in your category that don’t mention you yet, and what each one already cites.
- How to answer a campaignTake an investigated burst from Campaign Radar into the Strategist and work up your response to it.
- Which signals to build firstYour benchmark shows where you’re behind on links, coverage and branded search. The Strategist helps you pick which one to work on rather than trying to move all three.
It plans and drafts with you. Approaching people, and the relationships that come out of it, stay yours.
What to do with a gap once you can see it.
- Ask the Strategic AdvisorBring a visibility gap into the Strategic Advisor to work out why it’s happening and whether it’s worth doing before the other work in your queue. It reads your AI runs alongside your search performance rather than in isolation.
- Brief it in Content StudioSend a finding to Content Studio with the questions, the answers and the competing sources already attached, so the brief starts from the evidence.
- Track what happened nextMark the work as applied and Impact Tracker records the baseline from that date, so you can see what happens to mentions, citations and visibility afterwards.
Not connected yet? The free AI visibility check puts the same six assistants to work on your brand and we read the results with you.
Find out what AI is telling your buyers.
Connect your site and Searchscope tracks the questions that matter in your category across six assistants, week after week, and keeps every answer so you can read what was actually said.