Search Console archive

Google Search Console historical data, kept past sixteen months

Connect your Google account, run one backfill, and every month after that is yours for good.

The problem

Why Google Search Console historical data disappears

Sixteen months, then the window slides and the oldest month is gone.

Google holds sixteen months and no more. On the first of every month the far end of that window falls off, and nothing you do inside Search Console brings it back. Pebble copies it out before that happens, so month seventeen still exists somewhere. That is the whole product in one sentence.

Connect

Connect once, then sync when it suits you

Read only access, and nothing wider.

Pebble asks Google for the read only Search Console scope and nothing else. Once connected, Sync recent pulls the current month plus the three complete months before it. Full backfill walks the sixteen months Google still holds, one month at a time, recording progress as it goes, so if it stops you resume from where it stopped rather than starting again.

What is kept

What your Google Search Console historical data holds

The detail, not a summary of the detail.

Clicks, impressions, CTR and impression weighted average position, by query, by page and by day, kept until you delete it. Not twelve months, not twenty four. Everything the archive pulls is stored at the same grain Google served it at, so a year from now the row is still a row.

One month, four metrics
Clicks
Impressions
CTR
Position
The maths

Positions weighted by impressions, the way Google does it

A plain average across rows is wrong, and most tools do it anyway.

Every average position in Pebble is weighted by impressions, at every level of aggregation. It is the difference between a number you can put in front of a client and a number that quietly flatters you.

Coverage

You always know what you hold

A gap you can see is a gap you can fill.

A strip of month chips per property shows where you hold full query data, where you have daily totals only, and where there is nothing yet. Re importing a month replaces it rather than duplicating it, so a re run is always safe.

Coverage by month
Query dataDaily totals onlyNothing yet
Older history

A Google Search Console alternative for anything older

Google will not serve it, so you bring it.

Beyond sixteen months Google's API simply returns nothing, so upload a Search Console export instead, as csv or xlsx, one calendar month at a time and as many files at once as you like. There is also an ingest API with a token per workspace for anyone who would rather automate it. Both paths upsert, so nothing ever doubles up and your Google Search Console historical data grows backwards as well as forwards.

Freshness

The newest two days are never shown

Because Google's own newest days move under you.

Nothing newer than two days ago is imported or displayed. Google's most recent days are incomplete and revise upward for a while, and a chart that changes retroactively is a chart nobody trusts.

Honest about it

Three things worth knowing before you start

Stated up front, because you will find them anyway.

  • Query level clicks total roughly seventy percent of site level clicks, because Google withholds rare search terms. That is Google's privacy threshold, not a gap in the import.
  • Nothing syncs on a timer. Every source pulls when you press sync, so the archive grows when you say so.
  • A Search Console file export gives query data as one lump for the whole period with no dates attached, so query level history means one calendar month per file. We do not guess which month a query belongs to, because a guess in an archive is worse than a gap.

Your data is being deleted right now.

Google's 16 month window is rolling forward as you read this. The earliest month you can still save is the one that disappears next.

Start archiving free