Recent advertising results are often incomplete. Customers may not have finished the buying process, and completed outcomes may not yet be reflected in every report.
Those are different delays. Separate them before deciding that yesterday’s spending suddenly became less effective.
Distinguish behavior from processing
Conversion lag is the time between the relevant ad interaction and the customer action. Reporting delay is the time before data appears or settles in a reporting system. Offline workflows can add their own upload timing.
Google provides guidance for examining how long customers take to convert. Use current account evidence rather than assuming every campaign has the same lag.
Compare like with like
A mature historical cohort has had more time to produce outcomes than a cohort acquired this week. Comparing their current acquisition costs without accounting for that difference can create a false deterioration.
For a hypothetical sales process, many customers might decide several weeks after the first enquiry. The most recent leads are unresolved, not necessarily failures. Keep the unresolved category visible.
Check for an actual implementation issue
Timing is not an excuse to ignore broken tracking. Follow known customer actions through the website, CRM, and relevant reporting path. Check whether identifiers, event definitions, and imports are functioning.
If the number stops changing after a known implementation change, investigate that event rather than simply waiting longer.
Build a maturity-aware report
Show the acquisition period, elapsed time, known outcomes, and unresolved opportunities. Where useful, compare cohorts at the same age. Keep the time basis explicit: interaction date, event date, and upload date may answer different questions.
Avoid presenting a precise expected final result unless you have a defensible model and enough history to support it.
Decide what can be changed now
Repair clear delivery or measurement failures promptly. For uncertain commercial outcomes, use the planned review window and budget limit. If the delay is too long for the available cash, change the test design rather than pretending the evidence will arrive sooner.
The small-sample decision guide addresses the related problem of low volume. A useful performance review makes incompleteness visible before it turns into an unnecessary campaign change.
Co-founder and CEO of Stackmatix, startup advisor, and former Head of Sales at MightyHive. Connect with Matt on LinkedIn ↗ · More about Matt →