History — every scan kept, every number exportable
Each scan is a dated row: the share of answers that named you, the share that cited your site, how many questions were tracked and how many sources were found. Above the table, the same series charted — named or cited, with the margin drawn as a band — so a move you can see is a move that cleared the noise. Every row opens its dossier; every number exports.
How it works
Same questions, same method, every week
Because the question set and the run count are fixed between scans, the series is like-for-like. Edit a question and its history stays with the questions you kept.
The band is the margin
The chart draws each week's rate inside its margin. Two points whose bands overlap did not move; the chart does not pretend otherwise.
Export everything
On Growth and above, one button downloads a CSV of every number behind the project: each question, each engine, each run, named and cited flags, the sources. For the spreadsheet, the client report or the audit of us.
In depth
Failures shrink the sample, visibly
When an engine call fails after retries, the missing answer is counted and shown. A rate over 112 of 120 planned answers says so, in the row and in the export.
Dossiers accumulate
Every row links to its dossier, so last month's report is one click away and reads exactly as it did then. Nothing is re-baselined.
What we won't draw: a daily line. Scans are weekly because at these sample sizes a daily point is noise with a date on it. A calmer chart is the accurate one.
Common questions
How far back does history go?+
To the project's first scan. Every scan is kept with its full answers, so any week can be re-read, not just re-plotted.
What is in the CSV export?+
Every answer cell: question, engine, run, whether you were named, whether your site was cited, the sources cited, and the scan-level rates with margins. Growth and Scale plans.
See it on your own domain — first dossier in about a minute.