Performance
Measuring native advertising performance
Measure native campaign delivery, arrivals, reading and qualified actions against one stated goal, with reporting limits made clear.
Measure a native campaign against the action it was meant to produce. Use delivery and post-click measures to explain the result: impressions show reported delivery, clicks show a response to the promotion, and recorded visits show an arrival under the analytics tool’s counting rules. None establishes a useful read or a business outcome on its own.
Choose the outcome before launch
Write one primary question for the report. Did the campaign bring visitors to an explanation? Generate eligible enquiries? Contribute to purchases within a defined period? Choose the measure that answers it, then record the supporting measures needed to diagnose a weak result.
| Stage | Measure to define | Question it answers |
|---|---|---|
| Delivery | Reported impressions by agreed surface | Did the promotion run where and when expected? |
| Response | Ad clicks | Did people act on the promotion? |
| Arrival | Tagged landing sessions or another stated visit measure | Did analytics record an arrival? |
| Reading | A defined action on the destination article | Did visits show evidence consistent with reading? |
| Outcome | An eligible enquiry, purchase or other specified action | Did the campaign advance its goal? |
For an educational article, a qualifying article visit may be the primary outcome. For an acquisition campaign, an eligible action matters more than a broad engagement rate. Fix the definitions before reviewing results so a strong click figure does not replace the original goal.
Five stages of a native advertising read
- DeliveryReported impressions by agreed surface — did the promotion run where and when expected?
- ResponseAd clicks — did people act on the promotion?
- ArrivalTagged landing sessions or another stated visit measure — did analytics record an arrival?
- ReadingA defined action on the destination article — did visits show evidence consistent with reading?
- OutcomeAn eligible enquiry, purchase or other specified action — did the campaign advance its goal?
Make the report comparable
Use consistent campaign parameters on destination URLs. Record the destination, campaign identifiers, creative versions, dates, Australian geography and spend basis. If costs are compared, use the same GST treatment and say whether production and measurement are included.
Outbrain records a click whenever a user clicks a content recommendation on a publisher site. Analytics tools record visits or sessions under their own counting rules. A user who arrives through Outbrain more than once may therefore be counted differently in the two reports.
A click still counts in Outbrain if a user navigates away before the page fully loads and the analytics tracking code executes. A bookmarked return may be recorded differently by Outbrain and an analytics tool. Browser privacy preferences, ad blockers or disabled cookies, images or JavaScript can make visits untrackable.
Differences in reporting time zones can affect daily totals. Outbrain uses referrer URLs to track clicks and conversions, so a click may not be recorded when a user leaves HTTPS and enters HTTP.
Request the breakdown needed for the decision, where the supplier provides it. Campaign totals may answer whether the overall goal was reached; source, creative and device views may help locate a shortfall.
Check implementation before judging the campaign
Compare the direction of results over the same dates before treating a gap as evidence about the audience. Check that analytics tracking is implemented consistently on the relevant destination pages, and record any remaining implementation limits beside the campaign result.
Interpret reading signals carefully
GA4 provides engagement signals, but a session meeting an engagement measure does not establish that the promoted article was read. No engagement signal proves comprehension.
If reading matters, define a qualifying action on the article and label its unit: visits, sessions or users. Use the same rule across the sources being compared. Inspect the destination and the measurement setup when a rate looks weak; the cause could be the audience, the ad’s promise, the page or tracking.
Keep conversion timing visible
Name the intended action and how long after an eligible interaction the business will assess it. Keep that observation period separate from a platform or analytics attribution lookback window, which determines which earlier interactions can receive reported credit.
Outcomes from late campaign interactions may still be pending when delivery ends. Say whether a report groups attributed outcomes by interaction date or action date before reconciling totals.
Check the quality of an enquiry against the agreed eligibility rule. Attribution assigns credit under a reporting method; it does not establish that the ad alone caused the action.
Record the attribution method and settings used for the campaign report so the results can be interpreted consistently.
Observation period versus attribution lookback window
- Observation periodHow long after an eligible interaction the business will assess the intended action — set by the business.
- Attribution lookback windowA platform or analytics setting that determines which earlier interactions can receive reported credit.
- Grouped by interaction dateAttributed outcomes counted against when the interaction occurred.
- Grouped by action dateAttributed outcomes counted against when the action occurred. State which grouping is used before reconciling totals, because outcomes from late campaign interactions may still be pending when delivery ends.
- Quality checkCheck each enquiry against the agreed eligibility rule. Attribution assigns credit under a reporting method; it does not establish that the ad alone caused the action.
End with a decision
Report the primary outcome, supporting counts, period, cost basis and material measurement gaps together. Then decide whether to continue, repair tracking or the destination, change distribution, or stop. Tie the decision to the stage where the evidence shows a problem.
Close the report with a decision
- Report the primary outcomeThe one measure chosen before launch to answer the campaign's question.
- Include the supporting countsDelivery, response, arrival and reading measures that explain the result.
- State the periodThe dates covered and the observation period applied to outcomes.
- State the cost basisThe same GST treatment across campaigns, and whether production and measurement are included.
- Name the material measurement gapsTracking limits, untrackable arrivals and any remaining implementation constraints.
- Then decideContinue, repair tracking or the destination, change distribution, or stop — tied to the stage where the evidence shows the problem.
In this guide
- Measuring engaged reading after a native ad clickDefine and check qualifying article visits after a native ad click without treating GA4 engagement or scroll as proof of a complete read.
- Distinguishing cheap clicks from useful readershipCompare low-cost native clicks with recorded arrivals and qualifying article visits before judging whether a traffic source brings useful readers.
- Evaluating native campaign conversions over a suitable windowChoose a business observation period, distinguish it from attribution lookback and compare native campaign conversions after equal follow-up.
