Prompt
How do I measure how often my brand shows up in ChatGPT answers?
Latest observation
To measure how often your brand shows up in ChatGPT answers, you need to turn it into a repeatable brand visibility audit. The basic idea is:
- Ask ChatGPT a set of relevant questions.
- Check whether your brand appears in the responses.
- Track the percentage over time.
Here’s a practical way to do it.
1) Define the questions you care about
Create a list of prompts people might ask when looking for products like yours.
Examples:
- “What are the best CRM tools for small businesses?”
- “What’s a good project management app for remote teams?”
- “Recommend budget-friendly running shoes.”
Use a mix of:
- Generic category questions
- Comparison questions
- Use-case questions
- Problem/solution questions
The more your prompts reflect real user intent, the more useful your measurement will be.
2) Run the prompts consistently
For each prompt, capture:
- The full ChatGPT answer
- Whether your brand is mentioned
- Where it appears in the answer
- Whether it’s recommended, compared, or just mentioned
You can do this:
- Manually for a small set of prompts
- With automation using the OpenAI API or another LLM evaluation workflow
3) Measure simple metrics
A few useful metrics:
Brand mention rate
% of answers where your brand appears at least once
Formula:
(answers mentioning your brand / total answers) × 100
Recommendation rate
% of answers where your brand is actively recommended
This is stronger than a mention because it reflects positive visibility.
Rank/position
If ChatGPT lists multiple brands:
- First mention
- Top 3 inclusion
- Ordered ranking position
Share of voice
If multiple competitors are included, measure:
- How often your brand appears compared to others
- Your share of all brand mentions
4) Compare against competitors
Brand visibility matters more when benchmarked. Track:
- Your mention rate vs competitors
- Your recommendation rate vs competitors
- Prompts where competitors appear but you don’t
This helps you identify gaps in category coverage or positioning.
5) Segment by prompt type
Break the results down by:
- Product category
- Industry
- Use case
- Geographic market
- User sophistication level
Example:
- High visibility in “best tool for startups”
- Low visibility in “enterprise alternatives”
That tells you where to improve content, SEO, or brand presence.
6) Repeat over time
Run the same prompts weekly or monthly so you can track trends.
Important: model outputs can vary, so use:
- The same prompt set
- The same evaluation method
- Multiple runs per prompt if you want more stability
7) Use a scoring template
A simple spreadsheet might include:
| Prompt | Brand mentioned? | Recommended? | Position | Competitors mentioned | Notes |
|---|---|---|---|---|---|
| Best CRM for small business | Yes | Yes | 2 | HubSpot, Pipedrive | Listed in top 3 |
| Affordable email marketing tools | No | No | — | Mailchimp, Brevo | Not visible |
8) Important caveats
ChatGPT answers are:
- Not fully deterministic
- Influenced by prompt wording
- Sometimes incomplete or outdated
- Not the same as search engine rankings
So treat this as a visibility signal, not a perfect truth source.
Best practice
If you want this to be reliable, build an “LLM brand tracker” with:
- A fixed prompt set
- Automated API queries
- A consistent scoring rubric
- Competitor benchmarking
- Trend reporting over time
If you want, I can help you create:
- a Google Sheets template for tracking this,
- a Python script to automate it, or
- a scoring framework for brand mentions vs recommendations.