Scans

A scan is one execution of all your active prompts across the AI models your plan supports. Scans produce the data you see everywhere else in the dashboard.

How scans work

For each scan, SimplyRank:

1. Sends every active prompt to each supported AI model.
2. Captures the AI's response.
3. Looks for mentions of your brand and competitors.
4. Records visibility, position, and sentiment for each mention.
5. Extracts cited sources.
6. Aggregates everything into the visibility, trend, and competitor views.

On the Starter plan, for example, 50 prompts across 4 models is 200 individual queries in a single scan, which takes about a quarter of an hour. Bigger plans take longer, up to an hour or more. The first scan you run while setting up is a small set of trial questions and usually finishes in 2–8 minutes.

Scheduled scans

Most data comes from scheduled scans that run automatically based on your plan:

You set the frequency in Settings → Organization. Scheduled scans run at roughly the same time each day or week so trends stay comparable.

Manual scans

Owners, admins and members can run a manual scan from Scans when their plan and remaining allowance permit. Viewers can read results but cannot start scans. The free trial includes one sample scan; after it is used, choose a plan to scan again.

1. Open Scans in the sidebar, or Settings → Scans.
2. Click Run New Scan to scan your active prompts.
3. Follow the progress in the scan timeline.

To scan one question, open Prompts and use Scan Now on its row. Manage prompts on Scans opens the same prompt list.

On paid plans, manual scans count toward your monthly query limit, same as scheduled ones. Use them when:

• You've just added new prompts and want results today.
• A competitor launched something and you want to see if AI's mentioning it.
• You're testing a content change and want a faster signal than waiting for the weekly scan.

Scan history

The Scans page shows your five most recent scan sessions:

• Status — completed, in progress, failed, stale.
• Duration — how long it ran.
• Prompt coverage — what % of prompts were successfully processed.
• Models — which AI models the scan covered.

Viewing your results

Click View results on a completed scan to open the saved answers from that specific run. Each entry shows the prompt, AI engine and whether your brand was mentioned. Expand View AI answer to read the full response. The range above and below the list shows how many you are viewing, for example, Showing 1–25 of 40 saved answers (illustrative). Use Next answers and Previous answers at either end of the list.

Each expanded answer lists its Saved sources as clickable links. Original reference numbers were not saved, so this list does not claim to map each numbered marker to a particular page. If no source links were saved, the answer says so.

A failed or cancelled scan with saved answers also offers View results, clearly marked as incomplete. These are saved responses for inspection; incomplete scans may be excluded from dashboard analytics. Prompt wording on this screen reflects its current saved text, which may have been edited since the scan ran.

For summary metrics, open Overview. For visibility over time or comparisons by model, term or category, open Tracking. These analytics combine eligible scans for the selected brand and date range; they are not a report of one particular run.

A message saying Scan started means the scan is still processing. It is only described as completed when the service reports completion. If no scan started, check the message for an inactive prompt, used trial scan, monthly allowance or another refusal.

Statuses explained

If a scan stays stuck or repeatedly fails, see troubleshooting.

Query limits

Each scan consumes queries (prompts × models per scan). Paid plans cap how many queries you can run per month; the trial includes one sample scan:

On paid plans, if you use up your monthly allowance, scheduled and manual scans pause until the count resets at 00:00 UTC on the 1st of the month. The Scans page and Settings → Billing show how many queries you've used and, once scans pause, when they start again. On a paid plan that hasn't been cancelled, whoever created the workspace also gets an email when a scheduled scan is skipped, at most once a month. Upgrade or wait — see Changing your plan. After the trial sample is used, choose a paid plan to scan again.

What counts as a query

One prompt sent to one AI model. Two worked examples:

• Starter — 50 prompts × 4 models = 200 queries per scan, and running that daily for 30 days is 6,000 queries a month.
• Pro — the prompt allowance triples to 150, so 150 × 4 = 600 queries per scan and 18,000 a month.

Both land exactly on the plan cap, which is how the caps were set in the first place. Weekly scans work the same way. For example, on Starter Lite 25 prompts × 4 models = 100 queries per scan, and a month holds at most five weekly scans, so its cap is 500 queries a month.

What's next

• Trends and history
• Plans and pricing
• Common issues — scan stuck or failed