Data freshness & caching
Why the dashboard isn't live to the second, and how to make sure you're looking at current data.
The dashboard is fast because it doesn't re-fetch every source on every click. Instead, your data is pulled on a schedule and cached in between. Here's what that means in practice.
Why caching
Pulling live from every platform on each page view would be slow and would burn through each provider's API limits. Caching keeps the dashboard snappy and your integrations healthy, while still keeping data current enough to act on.
How current is "current"?
- Data refreshes on your chosen cadence (typically weekly or daily).
- Each period (7 / 30 / 90 days) is cached separately, so switching periods is instant.
- You can Refresh now for an immediate update, subject to a short cooldown.
The freshness chip on each tab
Every platform tab (Google Ads, Google Analytics, Search Console, Meta Ads, Meta Pages, Instagram, TikTok, LinkedIn and X) shows a small freshness chip at the top, so you always know how recent that tab's numbers are:
- When current: "Updated {time} · next refresh {when}" - how recently the data was pulled and when the next scheduled refresh is due.
- When overdue: an amber "Data may be outdated · updated {date}".
Staleness is judged against your cadence, so a daily tab is flagged sooner than a weekly one, and a manual tab is judged only against its last pull.
Next to the chip you'll also get a one-click Refresh now button that pulls fresh data straight from the connected platforms. It's rate-limited to once per hour, so if you've just refreshed you'll be asked to wait before pulling again.
On agency-managed accounts the chip is read-only - you see when the data was last updated and when it's next due, and your agency plus the automatic schedule keep it current. Team members get a Refresh now button next to the chip (once per hour) for an immediate pull.
Making sure you're current
- Check the freshness chip at the top of each tab (or the last-refresh time on the dashboard / management page).
- If it's overdue, run a manual refresh.
- After connecting a new platform, its data appears once the next refresh completes - it isn't instant.
When a source is briefly stale
If a platform can't be reached during a refresh, the rest still update and the gap is flagged - your existing data isn't wiped. CRM data that hasn't synced in a while is marked as possibly stale so you know to refresh.
When a source stops altogether
A one-off blip resolves itself on the next refresh. A connection that has genuinely stopped - an expired login, revoked access, a platform we can no longer reach - will not, and that tab would otherwise sit on old figures with only the freshness chip to show for it.
If you have Slack alerts connected, we tell you. You get one message naming the platform when it stops, and one when it starts working again - not a repeat every day it stays down. Connection alerts are sent as warnings, so they're switched off if you've set your minimum severity to critical only.
Without Slack connected, the freshness chip on the tab is the signal to watch: if a single platform is stuck on an old date while the others are current, that connection needs attention.
A paused connection is flagged on your Connections tab. If we hold a platform back because it stopped returning data we could trust, a note appears there naming it and explaining what happens next. Your existing figures stay on screen in the meantime.
The freshness chip is the other signal: if a single platform is stuck on an old date while the others are current, that connection needs attention.
Slack alerts announce a connection dropping and recovering as they happen, and reach any client - managed included - who has connected Slack. Clients without it rely on the Connections-tab note above.
Either way the fix starts the same place - open Connections, press Test connection on that platform, and it will tell you whether it needs re-authorising on your side or is something for us to fix.
To control cadence and run refreshes, see Data refresh & cadence.