Campaign use cases
Planning a Traffic Filtering Service Around Real Decisions
A filtering service is useful when it turns mixed traffic into decisions that can be predicted, inspected, and corrected.
Describe unwanted behavior precisely
Replace labels such as 'bad traffic' with observable conditions: an excluded country, excessive repeat activity, automation traits, a disallowed network category, or a missing campaign source. Precise definitions prevent unrelated visitors from being grouped together.
Connect classifications to policy
Decide which conditions route conservatively, which are merely recorded, and which trusted cases override a broader rule. The destination is a business decision; signal collection alone should not determine it without a documented mapping.
Create source-specific flows
Keep campaigns with different acquisition patterns separate enough to compare their event mix. Begin with a small set of high-confidence controls, validate them, and add combinations only when the logs expose a real gap.
Avoid permanent set-and-forget filters
A rule tuned during one traffic mix may become overly broad after a placement or regional expansion. Hidden defaults and empty referrers can also surprise teams that assume every request arrives with complete browser context.
Review outcomes at two levels
At the event level, confirm the reason matches the request attributes. At the campaign level, compare route distribution and error reports by source before and after a change, while investigating legitimate sessions that landed on the conservative page.