Every agent team runs in three modes — once, manual, autonomous — and the third one is really just this chapter: a schedule, plus the loop closed. Here's the scheduling layer in one place.
The two schedule families
| Schedule | What it drives |
|---|---|
| Data sync | The daily pull of Search Console / Analytics / store data into your dashboards — the fuel for everything else |
| Agent runs | Automatic SEO/optimization runs after each sync, research sweeps refilling your brief queue, and any team you've set to keep going |
Both live in the relevant team's settings; the sync schedule sits with your domains. Runs triggered by a schedule look exactly like runs you started by hand — same reports, same history, same credits accounting.
Times and timezones, without surprises
You enter times in your browser's local clock, and the interface shows which timezone it detected right under the field. Storage and execution are UTC underneath — meaning a 6:00 schedule set in Berlin runs at your 6:00, and daylight-saving transitions are handled at entry time rather than silently shifting your runs an hour.
Choosing a cadence
- Sync daily. Search data ships with a delay measured in days; more than daily buys nothing.
- Optimization after sync — the natural trigger, so runs always see fresh numbers.
- Research weekly-ish — briefs age slowly; a weekly sweep keeps the queue warm without flooding it.
- Marketing and ads: no schedule. Deliberately — drafts generate on demand, because nothing posts or spends on a timer.
Stopping is a toggle. Autonomous mode isn't a commitment — flip the schedule off and the loop pauses where it stands; everything already produced stays.
← All posts