Media measurement planning

A media measurement plan states the result a campaign should influence, what each channel can show, and how the combined result will be judged. Write it before activity starts, so convenient reporting figures do not become the default definition of success.

Start with the decision

Name the decision the evidence must support: continue the campaign, change its message, adjust investment or investigate a weak result. Choose a primary outcome that fits that decision, such as completed sales, qualified enquiries or bookings. Define which records count, the Australian market or service area, the reporting period and the source of the business record.

A campaign may also have an intermediate objective. If customers are unlikely to buy during the campaign, a suitably designed awareness survey may show progress. Give that measure its own definition and method; do not present it as a sales result.

Give each measure a job

Keep definitions and periods beside the numbers. An impression does not establish that someone saw an ad. A purchase credited to an ad under an attribution rule does not, by itself, establish that the ad caused it.

Traditional measures such as reach, frequency and TARPs can help describe media delivery, but may not show whether advertising achieved a business outcome. Keep them in their delivery or audience role rather than using them as substitutes for the outcome measure.

The layers help diagnose a result: poor delivery calls for a different investigation from adequate delivery followed by weak qualified demand.

Document data limitations

Ask each data owner to explain the source and method behind its figures, including changes that affect continuity. This helps distinguish a genuine change in campaign performance from a change in the data or reporting process.

Set expectations for what cross-channel reporting can and cannot establish. State any limits on linking exposure, channel activity and the business result.

The measurement toolkit is maturing, but methods need to work together. Keep each method's role visible in the plan rather than combining outputs as if they answered the same question. A method may describe audience, delivery, response or contribution; these are not interchangeable.

Key Data Sources and Guidance

Plan for a combined campaign view

Use one business outcome definition across channels, even when channels report different supporting metrics. Decide how duplicate orders or enquiries will be handled in the business total. Ask suppliers for their conversion definitions, counting methods, attribution windows and treatment of clicks and ad views. Their reports can then explain activity within each system without being added together as independent customers.

For each supplier, record whether credit can follow a click, an ad view or both, the attribution window for each, and the rule for deduplicating conversions. Treat those settings as definitions of reported credit, not proof that exposure caused the outcome.

Set a baseline and record other influences likely to affect the result, such as a price change, promotion or product shortage. Choose a comparison method that fits the question and available data.

A before-and-after view provides context but cannot, on its own, isolate media's effect. A well-designed experiment can help estimate an effect against a comparison; attribution reporting allocates credit to observed or modelled actions under stated rules.

Google Analytics attribution can favour lower-funnel activity such as brand search. A radio campaign may have influenced a later search 8 months before it was recorded, so a platform view may not capture the full cross-channel context.

Define shared terms and data boundaries

For audience and delivery measures, agree on the terms and data standards that make channel reports interpretable together.

IAB Australia identifies standardised, currency-grade audience data as the industry's common language, while cross-media audience measurement remains an unresolved gap. Record whether each channel's audience figures use shared definitions and state where the data does not support fair cross-channel comparison.

Do not assume that similar-looking figures are directly comparable. Media consumption is fragmented across platforms, formats and devices, and the loss of cross-platform identifiers and addressable signals makes it harder to link exposure to business outcomes or compare channels fairly.

Make data availability part of the plan, not a launch-day discovery. Record which datasets and signals each channel can provide, what can be matched across sources, and where gaps prevent a combined view.

Make the plan usable

Before launch, identify each data source's owner, collection method, reporting delay and fallback if data is missing. Assign an advertiser-side owner for the business outcome record and a supplier or platform contact for each channel report. Check whether the business outcome can be retrieved for the planned geography and dates.

If the design involves personal information or tracking technology, review the actual collection and sharing arrangements against the Privacy Act 1988 (Cth) and Australian Privacy Principles (APPs), where applicable. Consult the Office of the Australian Information Commissioner (OAIC) Guide to data analytics and the Australian Privacy Principles when checking analytics arrangements.

Agree on an early check for delivery and data quality and a later review that allows for the agreed response period. Keep dated records of changes to events, offers and attribution settings. In the final report, separate delivery, the business result, attributed activity and remaining uncertainty.

Check collaboration readiness

If the plan depends on data being combined between organisations or platforms, confirm in advance what can actually be shared and how the resulting view will be interpreted. Do not make privacy-safe data collaboration or clean rooms an assumed part of the plan.