AI visibility report template: what to send clients every month

A nine-section structure for a monthly AI visibility report clients understand — with example wording, what to leave out, and how to present uncertainty without losing trust.

Ranxii team4 min readTemplate
On this page
  1. The structure at a glance
  2. 1. Executive summary
  3. 2. Visibility this period
  4. 3. Change from previous period
  5. 4. By AI engine
  6. 5. Competitor share of voice
  7. 6. Strongest questions
  8. 7. Missing questions and opportunities
  9. 8. Most cited sources
  10. 9. Methodology
  11. What to leave out
  12. Presenting it
  13. Producing it every month without the busywork

A good AI visibility report answers four client questions in this order: Are we being recommended? Is it getting better? Who's recommended instead? What are you doing about it? Everything else is supporting evidence.

Below is the structure we use — nine short sections — with example wording you can adapt. It works as a standalone report or as a section inside your existing SEO report.

The structure at a glance

#SectionAnswers
1Executive summaryWhat happened, in three sentences
2Visibility this periodAre we being recommended?
3Change from previous periodIs it getting better?
4By AI engineWhere is it strong or weak?
5Competitor share of voiceWho's recommended instead?
6Strongest questionsWhat to defend
7Missing questions and opportunitiesWhere to act
8Most cited sourcesWhich websites shape the answers
9MethodologyWhy the client can trust the numbers

Your recommendations sit after section 8 or inside the summary — they're the agency's work, and the reason the client pays you rather than a tool.

1. Executive summary

Three sentences, written for someone who will read nothing else.

(Fictional client; the same data as our sample report.)

2. Visibility this period

The headline numbers, always with counts:

  • Mention rate: mentioned in 71 of 240 answers (30%).
  • Citation rate: the client's website cited as a source in 60 of 240 answers (25%).
  • Prompts covered: 20 buyer questions, asked weekly on ChatGPT, Gemini and Perplexity.

Definitions for each metric are in how to measure AI visibility.

3. Change from previous period

Same prompts, same engines, previous month. Show the change in answers, not just percentage points — "+17 answers mentioning Northwind" is more concrete than "+7 pts". List the questions gained (absent before, mentioned now) and lost. Call small changes what they are: within normal variation. How often to track AI visibility explains how big a change needs to be before it's real.

4. By AI engine

One line per engine with mention rate and citation rate. Engines differ a lot; the average hides it. Add one sentence of interpretation: "Strong on Perplexity, where the client's own site is often cited; weak on ChatGPT." Background in ChatGPT vs Gemini vs Perplexity.

5. Competitor share of voice

A simple bar chart or table: client and 3–5 competitors, each with its share of tracked mentions and its mention count. This is usually the section clients look at first. Counting rules matter; see how to calculate AI share of voice.

6. Strongest questions

The 3–5 prompts where the client is mentioned in most answers. These are positions to defend, and good news worth saying out loud.

7. Missing questions and opportunities

Prompts where the client was never mentioned while competitors were — with the competitors named. This is the most actionable section in the report and the natural bridge to your recommendations — why ChatGPT recommends a competitor covers how to diagnose each gap.

8. Most cited sources

The domains engines cited most often for these questions. Mark the client's own domain. Review sites, directories and comparison articles that appear repeatedly are where PR, partnerships and listings work pays off.

9. Methodology

Short, plain and always included. It's what makes the rest of the report defensible:

What to leave out

  • Rankings. "#2 in ChatGPT" will be contradicted the first time the client checks. Here's why.
  • Screenshots as evidence. Use one as an illustration if it helps; never as the number.
  • Unexplained scores. If you can't say how a 0–100 "visibility score" is calculated, don't put it in front of a client.
  • Every metric you have. Nine sections, one page each at most. If a chart doesn't change what the client does, cut it.

Presenting it

  • Lead with the summary and the competitor table in the meeting; keep the rest as reference.
  • Open two real answers — one where the client is recommended, one where a competitor is. It makes the numbers concrete.
  • End with three actions for next month, tied to section 7. If AI visibility isn't yet a paid service, here's how to sell it.

Producing it every month without the busywork

The slow part isn't writing the summary; it's collecting and counting hundreds of answers per client. Ranxii generates this report automatically each month — the same nine sections, with your agency's logo and colour — from the answers it stores every week. Share it as a private link that you can switch off at any time, or save it as a PDF, and add your recommendations on top.

Ranxii · AI visibility reporting for agencies

See the report your client would receive

Mention rate, change since last month, engine breakdown, competitor share of voice, missing prompts and cited sources — every number traceable to stored answers.