See exactly where you are absent. And what is winning instead.
Every topic you are missing from, broken out by engine, with the pages being cited in your place and the keywords they share. A visibility score tells you there is a problem. Gap Analysis hands you the work list.
Gap register
Open gaps this week
14
Open gaps
−18%
Week over week
5 / 2
Closed / opened
| Topic | Absent on | Weeks open | Status | History |
|---|---|---|---|---|
| Best trail running shoes for beginners | ChatGPTPerplexity | 6 | Recurring | |
| Running shoes for flat feet | ChatGPT | 3 | Recurring | |
| Carbon plate racing shoes compared | Google AI OverviewGemini | 1 | New | |
| How to choose running shoe size | Perplexity | 4 | Recurring |
Prompt evidence
Open one gap and the work is already scoped
Every prompt you are absent from carries its own evidence: the pages winning it, how hard each one is entrenched, and the terms all of them share.
Prompt
“best trail running shoes for beginners”
4
Pages beating you
6w
Open
What is being cited instead
Cited in / runs measuredfleetfeet.com
/shop/trail-running-shoes
runnerworld.com
/gear/best-trail-running-shoes-for-beginners
roadrunnersports.com
/guides/trail-shoe-finder
outsideonline.com
/outdoor-gear/run/trail-shoe-buying-guide
One of these is not a competitor. The Runner’s World roundup is cited on 7 of 8 runs and your shoes are not in it — that is one outreach email, not a content project.
What they all have
Shared keywordstrail running shoes
Table stakesbeginner trail shoes
heel-to-toe drop
waterproof trail shoes
Table stakes means every page beating you covers the term. Persistence weights it by how many runs those citations actually held — volume alone would rank a term nobody cites.
How gaps close
From absence to published, in three moves
Isolate the gap
Filter the register down to a single engine and topic. A gap on ChatGPT is a different problem from the same topic on Perplexity, and it usually needs different content.
Read the incumbents
Open the prompt to see the pages cited in your place, how often each recurs, and whether the citation is going to a competitor or to a third-party listing you could be included in.
Ship against the brief
Take the shared keywords, coverage counts, and page types straight into a content brief — then watch the entry’s weekly history to confirm the gap actually closed.
One gap = one topic × one engine. Absent on ChatGPT does not mean absent everywhere.
Recompute on demand. No waiting for the next scheduled cycle.
Trusted by growing brands
See how businesses are turning AI search visibility into real customers
“Seerly completely changed the game for us! We started getting actual paid customers from AI traffic, and it happened so much faster than we expected. The platform showed us exactly where we were invisible to AI engines and gave us a clear playbook to fix it. Within weeks, the results were tangible. Honestly, it's been a game-changer for our growth.”
Frequently asked questions
Everything you need to know about Seerly gap analysis
A gap is one topic paired with one AI engine your brand is absent on — so the same topic can be an open gap on ChatGPT while being perfectly healthy on Perplexity.
Treating the pair as the unit matters, because engines disagree. A single blended "visibility score" hides the fact that you are missing from one surface entirely, and it gives you nothing specific to fix.
- The topic the gap belongs to
- Which engines you are absent on, individually
- How many weeks the gap has stayed open
- Whether it is newly opened or recurring
Visibility tracking answers "how visible are we". Gap Analysis answers "exactly where are we absent, and what is winning there instead".
Visibility gives you the trend line. Gap Analysis gives you the work list: the specific prompts you are missing from, the pages being cited in your place, and the keywords those pages share. One is measurement, the other is a queue.
How citation tracking worksOpening a topic shows every prompt you are absent from, and for each prompt, the pages AI engines cite instead of you.
- The cited URL and its source domain
- The page type — product or service page, blog, homepage, listing, or social
- Whether that domain is a direct competitor, an indirect one, or a third-party source
- How many analysis runs and how many weeks it has kept being cited
Recurrence is the signal worth acting on. A page cited once is noise; a page cited on every run is the incumbent you actually have to displace.
Seerly extracts the keywords shared across the pages winning a prompt, so a gap arrives as a brief rather than an observation.
For each keyword you see its search volume, how persistently it appears across analysis runs, and how many of the cited pages actually cover it. A keyword carried by every winning page is a requirement for that topic, not a suggestion.
Turn briefs into contentThe register is tracked weekly, so you see the current number of open gaps, the week-over-week change, and how many gaps closed versus opened in that period.
Each entry also carries its own history, which is what separates a gap you have never addressed from one that keeps reopening after you publish. Those two problems need different responses.
No. Gap computation can be triggered on demand for a project, and progress is streamed live while it runs.
Repeat requests for the same domain coalesce into the running job rather than queueing duplicate work, so triggering it twice is harmless.
See analysis frequency by planReady to close the
gaps costing you answers?
See every topic and engine your brand is missing from, what is being cited instead, and the keywords you need to cover.