In short
AI visibility tracking runs a fixed set of buyer prompts across ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews, then reports how often your business is named or cited, versus competitors. Vatam delivers it monthly for businesses in Rwanda, Kenya, Uganda and Tanzania, alongside Search Console and analytics, as dated captures and trends, not guarantees.
What is AI visibility tracking?
AI visibility tracking measures how often AI assistants name or cite your business when buyers ask questions about what you sell. We run the same set of prompts on the same platforms every month and record the answers.
The platforms are ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews. The result is a series of dated snapshots you can compare, instead of a guess about what an assistant might say.
Why is a fixed prompt panel better than random checks?
Asking an assistant a question once tells you very little, because answers change by wording, user and day. A fixed panel keeps the questions the same, so any change you see comes from the answers and not from a different question.
We build the panel with you from the way real buyers ask. A Kigali boutique hotel might track “boutique hotel in Kigali” and “where to stay near Kigali Convention Centre”. A Nairobi fintech might track “best money transfer to Kenya”. A Dar es Salaam operator might track “best Serengeti safari operator” in English and in Swahili.
How is it different in East African markets?
Buyers ask in more than one language and from more than one country, and the same question can produce different answers depending on location. A panel for East Africa has to reflect that.
| Market | What the panel includes |
|---|---|
| Kigali, Rwanda | English and French prompts, plus visitor-style questions from international travellers |
| Nairobi, Kenya | Comparison and “best” prompts, which dominate the fintech and professional services space |
| Kampala, Uganda | Local-intent prompts for services and hospitality, in English and Swahili where relevant |
| Dar es Salaam, Tanzania | Swahili and English prompts, with safari and Zanzibar queries for international buyers |
What is citation share?
Citation share is the share of your tracked prompts where your business is named or cited. If you track a set of prompts across five platforms and you appear in a portion of the answers, that portion is your citation share.
We report it overall, by platform, by market and by prompt, and we show the same figure for the competitors you choose. It is a simple number, but we never present it without the underlying answers, so you can read exactly what each assistant said.
What do we do, step by step?
- Agree the panel. Prompts, languages, markets and competitors, signed off by you.
- Capture a baseline. The first run, usually from the AI Visibility Audit, becomes the starting point.
- Run it every month. The same prompts, the same platforms, dated and saved.
- Read the sources. We note which pages each assistant cited, because that tells you where to work next.
- Combine with search data. Search Console, analytics and any AI referral traffic that appears in your reports.
- Write it up. A plain-English report: what changed, why we think so, and what we will do next.
What do you receive and how do we measure it?
| What we do | What you get | How we measure |
|---|---|---|
| Prompt panel | A written panel of buyer prompts by market and language | Panel kept stable so months can be compared |
| Monthly capture | Saved answers from each platform, dated | Citation share overall, by platform and by prompt |
| Competitor comparison | Side-by-side view against the businesses you name | Change in each business’s citation share |
| Search and traffic review | Search Console and analytics summary | Clicks, impressions, queries and AI referrals where they show up |
| Written report | What changed, likely reasons and next actions | Whether the actions we took match the movement |
For the Lagos retailer in the LAGE Lightings case study, our published checks include Gemini naming the brand first for a buyer query, and ChatGPT and Bing referrals appearing in analytics. The Lens & Co case began with a July 2026 baseline of zero AI citations, and it relaunched in October 2026, and by October 2026 Google AI Overviews, ChatGPT and Claude were all mentioning the store, which tracking like this is designed to show.
What are the honest limits?
AI answers vary by user, location, wording and time. Two people asking the same question can see different results, so what we record is a dated capture, and we read trends over months, not single moves. Platforms also change how they work without warning.
Nobody can promise that a business will be named, and no tracking tool changes the answer on its own. Tracking tells you whether the work is moving the signals. Teams that want to learn to run a panel themselves can do so through training and workshops.
The free audit includes a first manual check across ChatGPT, Gemini and Perplexity. Request your free AI visibility audit to see your starting point.
Where does it sit in the packages?
The AI Visibility Audit is the first capture and baseline. Monthly tracking and reporting are part of the Growth retainer, because a trend needs several months of data. A Foundation build usually comes first when the site needs fixes before it can be measured fairly.
Tracking is the measurement layer for GEO and AEO, and it sits beside the classic reporting described under SEO. The industries page lists the sectors we work with.
Frequently asked questions
How do you track whether ChatGPT mentions my business?
We run a fixed list of buyer prompts on ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews each month, record whether your business is named or cited, and save the full answers with the date. Because the prompts stay the same, month-to-month changes are comparable.
What is citation share?
Citation share is the share of your tracked prompts where your business is named or cited by an AI assistant. We report it overall, by platform, by market and by prompt, and show competitors beside it. It is one useful number, always shown with the underlying answers.
Why do AI answers change from one check to the next?
AI assistants vary their answers by wording, user, location, time and model updates. That is why we treat each run as a dated capture and read trends across months. One check can mislead in either direction, so we do not draw conclusions from a single result.
Can you guarantee my citation share will rise?
No. No one controls what AI assistants say. We guarantee a measured baseline, a consistent panel, honest monthly reporting and the work most likely to move the signals. If a number goes down, the report says so and explains what we will change.
What is in the monthly report?
Citation share and prompt-level results, a competitor comparison, the sources AI assistants cited, Search Console and analytics highlights, and a short list of what changed and what happens next. It is written in plain English so you can read it on a phone without a glossary.