Skip to main content
Two widgets that look like they’re showing the same metric often turn out to be configured differently — even by a single filter or a widget-level date override. Work through the checks below before assuming the data is wrong.

​
Quick checks

1

Click Refresh data in the dashboard header

Re-runs every widget against the latest source data. Rules out cache disagreement between widgets.
2

Compare the widget settings side by side

Metric, date range, breakdowns, filters, data source list.
3

Confirm both widgets use the same date range

A widget-level date override that differs from the dashboard filter is the single most common cause.

​
Likely causes

A widget can override the dashboard-level date range from its own data settings. If one widget uses “Last 7 days” and the dashboard uses “Last 30 days,” the numbers won’t match. Disable the override to inherit the dashboard period.→ See Date ranges.
Filters apply at three layers — teamspace, dashboard, and widget — and any layer can drop rows silently. An extra widget-level filter on one of the two widgets is the second most common cause.→ See Filters.
Two metrics with similar names measure different things. Common examples:
  • Meta Lead vs Lead (Grouped) — Lead is the broader action_type=lead (includes offsite Pixel/CAPI leads plus on-Facebook leads); Lead (Grouped) matches Meta UI’s Leads (Form) column and covers Lead Ads instant forms only.
  • Google Ads Purchase vs Category: PURCHASE — the first is a single conversion action; the second aggregates all conversion actions categorized under PURCHASE.
  • Meta Purchases (Click, 7d) vs Purchases (Click, 28d) — different attribution windows.
  • Google Search Ads 360 conversion metrics with (conv time) vs their asterisked counterparts.
Custom metrics can also collide when a raw metric is renamed to the same name as a system metric (a renamed Meta Conversion Value → Revenue, for example) — the formula then resolves against the wrong input.→ See the Breakdowns & metrics FAQs for the common formula-scoping pitfalls.
Data blending uses a left join by default: the blended source is scoped to values that also appear in a marketing source. Two widgets built on the same blend but with different join types will produce different results. Full join contributes rows from the blended source of its own — and forces every mapped filter to match on both sides simultaneously.→ See Data integration FAQs.
A widget that includes one extra channel (Meta plus Google Ads vs. Meta only) will show a different total. Confirm the source list matches.
Widgets grouped at different levels (campaign vs. ad set, for example) don’t always sum the same way — rows without a label at the child level get dropped there but not at the parent. Custom breakdowns are especially prone to this: any campaign not caught by a label condition is excluded from the breakdown but still contributes to the ad-account total.→ See the Breakdowns & metrics FAQs.
Each widget caches its last fetched result in the browser. If one widget refetched recently and the other didn’t, the totals can briefly disagree. Refresh data in the header rebuilds both.

​
Still stuck?

Contact support

Email support@adriel.com with the dashboard URL and screenshots of both widgets’ settings panels.