What happens in a scan

A scan is Surfais asking your tracked prompts to five AI platforms, exactly as a real customer would, then reading the answers to see whether — and how — your brand shows up. This page walks through what happens between clicking “run” (or the schedule firing) and your dashboard updating.

​

The short version

Every scan runs the same pipeline, whether it’s triggered automatically on your plan’s schedule or manually from your dashboard:

​

The five platforms

By default a scan queries all five platforms Surfais tracks — ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews — for every prompt. Individual prompts can be narrowed to a subset of platforms from the prompt library, and the scan honours each prompt’s selection; platforms a prompt has switched off simply have no results for it. Keeping most prompts on all five is usually right — the AIS Score is built to reflect your visibility across the platforms your customers actually use, not just the one you happen to check yourself. See The 5 AI platforms we track for how each one is queried and why answers can differ between them.

​

What “per country” means

If you track more than one country, each prompt is asked once per country, with Surfais adjusting how it asks so the platform returns a locally relevant answer rather than a generic one. A UK-market prompt and a US-market prompt for the same question can — and often do — surface different sources and even different brands, because that’s what a real searcher in each market would see. Countries are configured in Countries & markets.

​

What gets stored from each answer

For every individual prompt/platform/country answer, Surfais keeps:

• Whether your brand was mentioned at all
• Its position relative to other brands named in the same answer (first mentioned, second, and so on)
• The sentiment of that mention
• Any competitor brands that appeared alongside yours
• The sources the platform cited, if any
• The raw answer text itself, so you can always read exactly what the AI said

This is the same underlying data that powers the Mentions feed, Citations & sources, and Sentiment monitoring pages — a scan doesn’t just produce a score, it produces evidence you can read prompt by prompt. See Mentions vs citations if the distinction between the two isn’t obvious yet.

​

While a scan is running

A scan typically takes 5–10 minutes to finish, start to finish, depending on how many prompts, platforms, and countries you’re tracking. You don’t need to sit and watch it — you can navigate anywhere else in Surfais, or close the tab entirely, and the scan keeps running in the background. While a scan is in flight:

• A banner appears across the app letting you know a scan is running for the active brand.
• Mentions land as they’re processed — so if you head to the Mentions feed mid-scan, you’ll see results filling in platform by platform, live.
• Your dashboard scores update only once the whole scan finishes. The AIS Score and its components are computed from the complete set of answers, not a partial one, so they deliberately hold steady until every platform has responded.

If this is your brand’s very first scan, the dashboard shows a dedicated “first scan in progress” view rather than an empty gauge, and tells you roughly when automatic scanning takes over afterwards. You’ll get an email as soon as results are ready either way.

​

When results appear

Once a scan completes, your dashboard refreshes with the new AIS Score, trend, and every metric beneath it. If this was your brand’s first-ever scan, you’ll see “Awaiting first scan” replaced with real numbers for the first time — that scan becomes the baseline every later reading is compared against. From here, Why scores move explains how to read changes between one scan and the next.

​

Related pages