Measurement
Part of Podcast ad frequency and reach
Reading campaign reach alongside response
Compare estimated podcast reach with recorded actions using aligned periods, clear definitions and honest attribution limits.
Treat estimated podcast reach and response as separate views of a campaign. Reach describes the scale of reported ad delivery; visits, enquiries and purchases describe actions recorded under their own rules. Neither figure alone shows how many people heard the ad or how many actions it caused.
Put the measures on one timeline
Show the campaign dates, eligible Australian audience, delivered impressions, estimated unique reach and average frequency. Beside them, state each response measure and its observation window. An order using a campaign code is an order recorded with that code; a dedicated-page visit is a visit recorded on that page. The same person may do both, so do not add the counts as if they were distinct responders.
Check that reach refers to the campaign ad rather than a show's audience. Ask how unique recipients were estimated and whether the reported total is deduplicated across all included shows. For open podcast downloads, server-side delivery does not confirm playback.
Investigate patterns without treating them as proof
| Observed pattern | Question to investigate |
|---|---|
| Reach rises, recorded response does not | Is the audience, offer, message or response path suitable? Is the observation window long enough? |
| Response rises, estimated reach is flat | Could repeat delivery, another channel, a shared code or ordinary demand contribute? |
| Both rise | What else happened at the same time, and what evidence supports a campaign effect? |
| Neither rises | Was the ad delivered as booked, and was response tracking working? |
These patterns suggest questions, not diagnoses. Dividing recorded actions by estimated reach produces a descriptive ratio, not a causal conversion rate: actions may come from outside the measured audience, and the reach denominator is estimated. Avoid comparing such ratios across sellers with different methods or windows.
Match evidence to the objective
For a direct action objective, place sales or lead-system outcomes beside code use, dedicated-page activity and any attribution report. State what each includes. A code can be shared beyond the podcast audience; someone influenced by an ad may act without using it. Matched impression-to-site reports depend on their tracking and attribution rules and do not count every action the campaign caused.
For a brand objective, consider a suitably designed study with exposed and comparison groups if budget and sample allow. Check how the groups were formed, what was asked and how uncertain the difference is. Awareness rising during a campaign does not, by itself, establish a podcast-caused lift.
Close with what the seller recorded as delivered, what response systems observed and what remains uncertain. Those three statements give a clearer basis for the next decision than a single return figure built from incompatible measures.
