The proportion of AI model prompts that include references to a brand’s assets.
Marketers read this in relation to overall AI traffic counts and analytics dashboards.
01What it is and how it works
When a large language model receives a user request, it scans the data it was trained on for relevant snippets. If a snippet contains a brand name, logo, or product description, that request counts as a prompt that features the brand. Prompt Share is calculated by dividing the number of brand‑featured prompts by the total number of prompts processed for a given period. The metric lives one level below the overall AI traffic count, giving marketers a view of brand presence inside the model’s reasoning.
It is the share of prompts that mention the brand.
02What to do about it
1. Pull the latest Prompt Share report from your analytics dashboard. 2. Identify the top‑performing prompts that mention your brand and note the surrounding context. 3. Add or update structured data (Schema.org Brand markup) on your webpages so the model can surface accurate, up‑to‑date information. 4. Create a short, factual brand description that can be used in prompt‑generation scripts. 5. Run a weekly audit of new content to ensure it follows the same tone and terminology. These steps can be completed within a single work week and will help you steer the model toward the messages you want.
03How it is measured or noticed
The platform logs every incoming prompt and tags it with any recognized brand entities. A separate counter tallies total prompts. Prompt Share = (brand‑tagged prompts ÷ total prompts) × 100. You can view the raw counts in the Prompt Share tab of the dashboard, export them as CSV, and plot trends over time. Look for spikes that coincide with marketing campaigns or news events; those are the moments when the model is most likely to surface your brand.
04Common mistakes
- Assuming a high Prompt Share automatically means higher sales – the metric only reflects mention frequency, not conversion.
- Counting duplicate prompts from the same user as separate instances – this inflates the share.
- Ignoring the surrounding context; a mention in a negative review still counts as a prompt.
- Treating Prompt Share like ad impression share; the underlying data sources differ completely.
05Limits
Prompt Share only applies to brands that are part of the model’s training corpus or that have been explicitly indexed by the provider. New brand assets that are behind paywalls or in private repositories will not be counted. The metric can be confused with search impression share, which measures visibility in traditional search results, not AI‑generated text. Also, Prompt Share does not capture brand mentions in images or audio generated by multimodal models.
06Worked example
A fashion retailer sees a Prompt Share of 2.4% for the month of June. The dashboard shows 12,000 brand‑tagged prompts out of 500,000 total prompts. The marketing team adds a concise brand description to their homepage and updates product schema. In July, the share rises to 3.1%, reflecting 15,500 brand‑tagged prompts out of 500,000 total.
User: "Show me sustainable sneakers similar to those from BrandX." AI response includes BrandX’s latest eco‑line.
Frequently asked questions
How does Prompt Share differ from a simple brand mention rate?
Usually, Prompt Share looks at the proportion of AI prompts that actually contain a brand reference, while a brand mention rate counts any occurrence of the brand in the model’s output. Prompt Share is focused on the input side, measuring how often the model is asked about the brand.
Should we try to increase Prompt Share for every brand in our portfolio?
It depends on the strategic value of each brand. If a brand is a core revenue driver, a higher Prompt Share can boost visibility, but for peripheral brands the effort may not justify the cost.
How is Prompt Share calculated on the platform?
Usually the platform logs every incoming prompt, tags any recognized brand entities, and then divides the number of prompts containing those tags by the total number of prompts processed in the period. The result is expressed as a percentage.
Does Prompt Share still work after we update our brand assets or add new keywords?
Usually it does, as long as the updated assets are indexed by the AI provider. If the new assets are not yet part of the model’s training corpus or indexing list, Prompt Share will not reflect them until the next update cycle.
What are the risks of a low Prompt Share for a brand?
It means the AI model rarely receives prompts that reference the brand, so the brand’s presence in AI‑generated content stays minimal. You may notice fewer organic mentions, lower brand recall in AI‑driven interactions, and missed opportunities for influence.
How long does it take for changes in brand indexing to show up in Prompt Share metrics?
Typically it takes a few days to a couple of weeks, depending on the provider’s refresh schedule. During that window you may still see the old baseline before the new data is incorporated.
Asked out loud
spoken, not typedThe same term in the words somebody uses speaking to an assistant rather than typing into a box — written from the situation, which is why each one carries the situation it came from.
Usually it does if the model has recent data that includes your brand. You can check the latest Prompt Share figure in the dashboard to confirm.
Yes, the platform tags any recognized brand assets in incoming prompts, so you can see whether they appear in the client‑facing responses. Look for the Prompt Share metric for that campaign.
It depends on whether the brand is part of the model’s training corpus or has been indexed. If it isn’t, the Prompt Share will be low and the brand won’t appear in the output.