Clicking through to a detail page should carry along the filters set on the dashboard, not reset them

What it is

When a user clicks a data point on the dashboard and lands on a detail page, the filters they set on the dashboard (time range, channel, region, device type, etc.) define the context of "what am I looking at." If the detail page resets to its default filters, the data scope differs from what the dashboard showed—the numbers on the two pages don't match, and the user either concludes the data is inconsistent or must manually reapply the filters. Carrying context through the jump is the precondition for a coherent "from detection to explanation" path.

Why it happens

Context loss stems from where the filter state lives: if filters exist only in front-end component memory and are not encoded into the URL or jump parameters, any navigation (to a detail page, or even a refresh) resets the state. The technical fix serializes the filter state into URL query parameters—then the detail-page URL naturally carries the context, and the link can be shared, bookmarked, and restored, solving handoff, sharing, and recovery in one move. The user-facing distortion is more insidious: a detail page that resets to defaults still "looks fine," and the user doesn't immediately notice the scope changed—they see a number without realizing it answers a different question. The distortion is especially dangerous for time ranges: the dashboard showed "3% anomaly rate over the last 7 days," the detail page defaults to "today" at 1.2%—both correct, neither comparable.

Where it stops holding

Not every filter should pass through: when the dashboard has mutually exclusive view modes (chart vs table), some filters may be meaningless on the detail page, and forcing them through creates confusion. There is also a counterintuitive boundary—context passing should be explicit: the detail page should display the inherited filters (in a breadcrumb or filter readout) so the user knows how the current data's scope was determined, rather than silently deciding for them. Security matters too: if the detail page contains sensitive data, URL-embedded context can leak into browser history and server logs.

Applying it

• Serialize all dashboard filters into URL query parameters; detail-page jump links inherit every parameter.
• Echo the received filters (time range, dimension values) at the top of the detail page, matching what the dashboard showed.
• For sensitive data, pass context via POST or a server-side session instead of the URL to avoid leakage.
• Verification: set non-default filters on the dashboard, click a data point to reach the detail page, and check every number's time range and filters against the source; any mismatch is a context handoff failure.

Related

• Same group: U7.08.1 An anomaly found on the dashboard needs a next station toward explanation · U7.08.3 The detail page must offer a way back that restores the dashboard's prior state · U7.08.4 A dashboard without drill-down paths can surface problems but not locate them · U7.08.5 Drillable elements need a clickable visual cue
• Nearby: U7.06.3 Relative and absolute ranges suit different usage scenarios · U7.05.5 The comparison period's statistical scope must match the current period's
• Search terms: deep linking · filter context · URL state