Six sources, one archive, and no surprises about whose quota is being spent
What each connection pulls, whose key it uses, and what it costs before it runs.
What connects, and on whose account
Read this as the whole answer, because there is nothing else running in the background.
Google Search Console
Pulls daily site totals, monthly query rows and monthly page rows. Sync recent covers the current month plus three complete months, full backfill walks Google's sixteen months and can be resumed.
Your own Google account, connected over OAuth with the read only scope.
Bing Webmaster Tools
Pulls site level daily figures and weekly query rows. Bing accepts no date range, so every sync returns the whole history Bing retains, roughly six months. Google and Bing figures are never added together anywhere.
Your own API key.
OtterlyAI
Pulls AI citations by month, by day and by engine, plus the prompts you appear in. A quota guard estimates the calls a date range needs and refuses to start a run it cannot finish.
Your own API key, owner only.
Semrush
Fills in volume, CPC, difficulty and intent for the keywords you track. A cost check runs first and tells you the exact number of API units before anything is spent, and blocks the run if your balance will not cover it.
Your own API key.
Keyword data
Volume, difficulty, intent, keyword ideas and the live SERP checks behind rank tracking. This is the only part of Pebble that spends our money instead of yours, which is why it carries a monthly allowance.
Our account, no key from you.
File upload
Search Console exports as csv or xlsx, Bing's CSV export, and Bing's AI grounding queries export. Everything upserts except the two snapshot datasets, which replace.
No credentials needed.
Ingest API
Accepts the same datasets as the uploads, for anyone who would rather automate it.
A token per workspace.
Your key, your quota, your data
Four of the six run on your credentials, on purpose.
Pebble does not resell anyone's data. Where a source is paid, you connect your own account and spend your own quota, which means you keep the relationship, the rate and the data. The one exception is our keyword and SERP data, which runs on our account and carries a stated monthly allowance so it can never quietly run away with itself.
Nothing runs on a schedule
Every sync is a button, and that is deliberate.
There is no background scheduler. You press sync on the Connections page and the data comes. When a connection spends your quota or our money, a job that fires by itself at three in the morning is a job you cannot supervise, so we did not build one. Coverage chips show you exactly what you hold, so it is always obvious when a sync is due.
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