Browsing perception analysis history
Compare past perception runs over time.
Last updated: 2026-05-05
Browsing perception analysis history
Every Brand Perception run is saved with a timestamp, the active filters, and a snapshot of the source scan data. The history popover lets you reload any past run and see the exact numbers as they appeared on that day — useful for board reports, regression checks, and proving impact after a content push.
Where to find history
On the Brand Perception page, the history popover sits in the top-right corner. It lists past runs in reverse chronological order with:
- Run timestamp
- Date range used
- Country
- Models included
- Composite sentiment score
- Number of gaps detected
Click any row to load that run.
Scope isolation
When you load a historical run, every panel on the page re-reads from that run's persisted data. Drivers, heatmap, gaps, recommendations, and the donut all reflect the same point-in-time snapshot. This prevents the most common reporting mistake — accidentally mixing today's drivers with last quarter's gaps.
The history badge at the top shows you're in historical mode and offers a one-click return to the live view.
Comparing runs
Open two browser tabs, load a different run in each, and compare side-by-side. (A native compare view is on the roadmap.) The composite score delta between two runs is a clean impact metric for stakeholder reports: "After the security content push in March, our security driver score moved from 0.42 to 0.71."
Retention
Historical runs are kept indefinitely on paid plans. Free plan retains the last 5 runs per brand.
Load history or re-run?
| You want to | Choose | Effect |
|---|---|---|
| Report on a past state | Load history | Every panel re-reads that run's persisted snapshot |
| Analyse the latest scan data | Re-run | Appends a new snapshot; history is preserved |
| Compare two points in time | Load each run in a separate tab | Composite score delta is your impact metric |
What history stores
| Item | Stored? | Note |
|---|---|---|
| Scores, gaps, drivers, attributes, recommendations | Yes | Kept indefinitely on paid plans; last 5 runs on Free |
| IDs of the underlying scan responses | Yes | Used for drill-downs |
| Raw response text | No | If a scan is deleted, drill-downs are unavailable |
| Active filters (date range, country, models) | Yes | Shown on each history row |
Re-running vs loading history
- Load history when you want to report on a past state.
- Re-run when you want a fresh analysis on the latest scan data. Re-running does not delete history — it appends a new snapshot.
What's stored, what isn't
Stored: every score, gap, driver, attribute, recommendation, and the IDs of the underlying scan responses.
Not stored: the raw response text. If a scan is later deleted, the perception run still loads but raw-text drill-downs are unavailable.
Related articles
- Brand Perception overview
- Perception scoring methodology
- Gap analysis
- Feature category heatmap