A metric that measures how many times an AI model selects a piece of a brand's content when responding to user queries.
Digital marketers reading analytics dashboards and reports on generative search impact.
01What it is and how it works
When an AI model like ChatGPT receives a user prompt, it searches its knowledge base and indexed web content to craft a response. Each time the model selects a piece of your brand's page, that interaction counts as one unit of Prompt Volume. The metric aggregates these selections across all prompts, regardless of the exact wording. It reflects the model's perceived relevance of your content to the user's intent.
Prompt Volume is the number of AI prompts that return your brand's content.
02What to do about it
Start by reviewing the top‑ranking prompts in your analytics dashboard. Identify the topics that repeatedly pull your content. Then: Update the corresponding pages with clearer headings and concise answers. Add structured data (FAQ, Article) to help the model surface the right snippet. Create new content that fills gaps you see in the prompt list. Set up alerts for sudden drops so you can react within a week.
03How it is measured or noticed
Our platform logs every API call where the model returns a URL belonging to your domain. The count is stored per day and can be filtered by prompt keyword, device, or geographic region. You can also see Prompt Volume in the “AI Search Impact” report, which shows a line chart of daily totals and a table of the highest‑volume prompts.
04Common mistakes
- Assuming a high volume means high conversion without checking click‑through rates.
- Optimizing only for the most frequent prompts and ignoring long‑tail queries.
- Removing brand mentions from content to lower volume, which defeats the purpose of visibility.
05Limits
Prompt Volume does not capture private or proprietary LLMs that do not query the public web. It also excludes prompts that trigger a generic answer without a direct citation to your URL. Do not confuse it with traditional search impressions; AI search can synthesize information from multiple sources, so a single prompt may count multiple times if your content appears in several answer fragments.
06Worked example
"User asked: 'How do I set up a sustainable coffee shop?' The AI included a link to our 'Sustainable Café Checklist' page. That interaction added one to our Prompt Volume for the keyword 'sustainable coffee shop setup'."
Frequently asked questions
How is Prompt Volume different from impression count?
It measures distinct AI queries that return your URLs, not the number of times a page is shown in search results. While impressions count every display, Prompt Volume only counts unique prompts that surface your content.
Should we start tracking Prompt Volume for all our brands?
Yes, if your marketing strategy relies on generative AI visibility, tracking Prompt Volume gives you insight into how often your brand is referenced. However, consider the effort required to integrate the data and whether your audience actually uses AI search.
How does the platform capture Prompt Volume data?
It logs each API call where the model includes a URL from your domain in its response. The system matches the returned link to your verified domains and increments the count for that prompt.
Does Prompt Volume still work now that some models use private knowledge bases?
It depends on whether the model queries the public web for answers. For models that rely solely on internal data, Prompt Volume will not capture those mentions, so the metric may underrepresent total AI exposure.
What happens if we misinterpret a high Prompt Volume?
You might assume strong brand relevance when the queries are unrelated or low‑quality, leading to misplaced investment. Look for the context of the prompts and the content they link to to avoid false confidence.
How long after a new piece of content is published does Prompt Volume reflect it?
Usually within a few days, as the AI model updates its index and starts serving the new URL in responses. In the meantime, monitor crawl logs and ensure the page is indexed.
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.
Yes, you can open the mobile dashboard and view the real‑time Prompt Volume metric for your brand. It shows the number of distinct AI queries that have returned your URLs in the last hour.
Usually you can ask the voice‑enabled version of the dashboard to read out the latest Prompt Volume for that page. It will tell you how many distinct AI prompts have surfaced the link today.
It depends; you should verify that the Prompt Volume counts are based on the latest API logs and that the URLs are correctly mapped to your domain. Cross‑check the raw log entries with the dashboard totals to ensure no data gaps.