4 signals in the AI monitoring report
Baseline window: agree the days covered before the first check. The report keeps 4 signals separate. Review what each answer showed, without relying on one blended score.
| Report field | What we log | What you can review |
|---|---|---|
| Mention | Brand or product appears | Answer wording and question |
| Citation | A source is visibly cited | Source address, when shown |
| Alternative | Another selected brand appears | Brands present in the same answer |
| Context | How the answer describes you | Use case, product details and accuracy |
Share of voice compares your appearances with selected alternatives across the agreed question set. It covers those checks, not every answer a person might receive. Each assistant gets a separate record. For the wider discipline, see AI search visibility (GEO).
The report makes it easier to spot the next useful check. A citation to an outdated page calls for a different review than an answer that names a competitor. We keep the example beside the summary, so your team can verify the observation.
How do we set up ChatGPT brand mention tracking?
3 choices shape a useful check: the buyer question, the assistant, and the comparison brands. Keep wording stable between reviews. Changed prompts make answers harder to compare.
We agree which product names, audience questions and alternatives belong in the first question set. Then we record answer context and visible citations for each selected surface. ChatGPT, Perplexity, Gemini, Microsoft Copilot and Google AI Overviews can be included when they matter to your audience.
- Questions: use real searches buyers ask before choosing.
- Names: include official brand and product spellings.
- Comparisons: identify alternatives you want logged.
- Surfaces: name each assistant or answer surface separately.
This gives ChatGPT brand mention tracking a repeatable basis. It does not assume every assistant displays sources in the same way. For a single-platform focus, see ChatGPT visibility or Perplexity optimization.
4 checks that make visibility data useful
4 checks turn raw answer observations into a decision aid. The first is coverage: did your brand appear for the questions you chose? The second is citation: did the answer visibly point to a source you own?
The third is comparison. Note which selected alternatives appear alongside your brand, using the same question set. The fourth is context: check whether the assistant describes the right product, audience and use case.
- Mention without citation: inspect how the answer frames the brand.
- Citation to your page: check whether the page supports the answer.
- Alternative in the response: review the question and answer context.
- No relevant mention: flag the question for closer review.
- Unclear product description: compare the answer with your official facts.
Read these signals together. A mention can matter without a visible citation. A citation can appear but point to a page needing attention. Check each example before deciding what to change.
For measurement choices and interpretation, see how to measure AI search visibility. For page changes, explore content for AI answers.
4 steps from first message to final report
4 steps take the review from your first message to a report with evidence. We confirm the question set before checks begin. We keep recorded fields consistent across the agreed assistants.
- Send the inputs: domain, product names, buyer questions, target audience and alternatives.
- Agree the checks: confirm question wording, selected assistants and report fields.
- Record the baseline: log answer context, brand mentions and visible citations.
- Review the findings: receive the summary, supporting examples and follow-up points.
You get a written quote within 2 working days. Work starts within 24 hours of confirmed payment. We confirm the baseline and recurring report cadence against your selected assistants and questions.
If you already use AI SEO monitoring software, send its export with your first message. We can compare its fields with this review. Then we agree which evidence your team still needs. A GEO audit or technical AEO review can follow when the report points to a site issue.
What can an AI visibility review confirm?
1 rule matters: the review records what appears in checked answers, not what every user will see. Assistants decide answer wording and whether citations appear. We cannot promise a brand mention, citation, or fixed share of voice.
That limit does not make the evidence vague. We preserve the question, answer context and visible source for each logged observation. The report also names the assistants included. Your team can see exactly what was checked.
When a pattern needs action, match it to the evidence. Review page accuracy when a citation points to weak or outdated content. Check product facts when an answer misstates your offer. Consider entity and knowledge graph work or digital PR for AI citations when those findings fit.
Send DigitalMedia1 your domain, product names, buyer questions and priority assistants. We’ll confirm the question set and share a written quote within 2 working days. After approval, we’ll start the review.
How it works
- 1. Send the inputsShare your domain, product names, audience, buyer questions and selected alternatives. Name the assistants you want reviewed.
- 2. Agree the checksWe confirm question wording, assistant coverage and the fields shown in the report.
- 3. Record the baselineWe log answer context, brand mentions and visible citations for the agreed checks.
- 4. Review the reportYou receive the summary with supporting examples and practical follow-up points.
Frequently asked questions
Which AI answers can you include in one review?
5 named surfaces can be considered: ChatGPT, Perplexity, Gemini, Microsoft Copilot and Google AI Overviews. We confirm the assistants that fit your audience before checks begin. Each gets a separate record. The report does not combine unlike answers into one unexplained result.
How do you compare share of voice across assistants?
1 agreed question set provides the comparison basis. We record when your brand and selected alternatives appear in those answers. Then we report observations by assistant. The comparison applies only to the questions and surfaces reviewed. It does not represent every answer users may receive.
What counts as an AI citation in the report?
1 citation is a source visibly shown with an answer during a check. We record the source when the interface exposes it. We distinguish that from a brand mention without a visible source. Your report keeps the cited page and answer context together for review.
Can you track brand mentions in ChatGPT and Gemini?
Yes. ChatGPT and Gemini can both be included, with separate question records for each surface. The same question wording helps make observations easier to compare. We also log visible citations and selected alternatives when those fields appear in the checked answer.
What does an AI visibility monitoring review cost?
Cost depends on 3 things: the surface, the question set and the reporting cadence. You get a fixed written quote within 2 working days, before you pay. Send your domain and priority assistants so we can confirm what the review will cover.
What should I send before the first monitoring check?
4 useful inputs are your domain, official brand and product names, buyer questions and priority assistants. Add the alternatives you want compared, plus any existing exports or screenshots. We use those materials as context. Then we agree the question set and report fields with you.
Tell us about your project
Answer four quick questions and a manager will send you a plan, timing and a price range within the hour. Everything stays confidential.
Loading the form…