Product capabilities

Using Click Logs and Live Statistics for Routing Operations

Live summaries show where to look, while individual click events provide the context needed to explain a routing decision.

Use aggregates and events for different jobs

Counts and ratios expose changes in volume, source mix, destinations, and reason clusters. An event record answers the narrower question by tying a timestamp, flow, request context, decision, and available evidence together.

Filter by the campaign's stable identifiers

Start with flow, source, campaign, placement, market, device class, network, and decision reason where available. Stable labels make comparisons repeatable and keep unrelated campaigns from distorting an investigation.

Define the time window precisely

Use a timezone and exact start and end points that match the launch, rule change, or reported incident. Compare an appropriate baseline window, accounting for scheduled source activity instead of interpreting every volume change as a routing problem.

Know what the log does not establish

Recorded headers and derived classifications describe the received request, not verified human identity or intent. Sampling, retention boundaries, client retries, preview requests, and tracker hops can also make event totals differ from other systems.

Turn an observation into a reproducible case

Save the filters, representative event identifiers, configuration version, expected behavior, and observed outcome. Re-run a controlled request after a change and verify that the intended segment moved without altering adjacent traffic.

Related guides