Click-based or view-based attribution for retail media, compared without picking a winner

Retail media attribution methods allocate sales credit under defined rules. For an England retailer comparing click-based and view-based reporting, hold the outcome population and reporting scope constant while making the different qualifying interactions explicit. Neither basis establishes causal incrementality by itself.

Comparison method and boundaries

This desk comparison is dated 6 September 2026. It concerns sponsored-product advertising on retailer or marketplace inventory. The alternatives are reporting bases, not ranked suppliers. No campaign data, paid account or observed accuracy test was available.

The common comparison unit is a credited qualifying order, with monetary totals excluded. Both alternatives must use the same property, product scope, eligible order cohort, reporting period, time zone, lookback duration, correction cut-off and order definition. Actual settings are UNKNOWN until documented. Different qualifying interactions remain an intentional difference, not evidence that the populations are identical in every respect.

Record event, denominator, cohort, period, time zone, attribution window, exclusions, corrections, exact source and uncertainty for each output. A credited-order count has no denominator; any derived rate requires a separately agreed eligible base. Exclude comparisons where these fields cannot be aligned or the assignment rules remain undocumented.

Compare the rules rather than the totals

The IAB Europe May 2026 standards distinguish click-based and view-based attribution reporting. This European methodology supplies comparison concepts, not an England effectiveness benchmark.

Criterion	Click-based reporting	View-based reporting
Qualifying evidence	Documented eligible click	Documented eligible view under the adopted definition
Outcome unit	Credited qualifying order	Credited qualifying order
Common scope and clock	Apply the shared specification	Apply the same specification
Credit precedence	Record actual rule; UNKNOWN here	Record actual rule; UNKNOWN here
Missing observation	Disclose click-record limits	Disclose view-measurement limits
Causal interpretation	Not established by credit alone	Not established by credit alone

Do not add the columns unless overlap and allocation have been resolved. Ask whether the same order can appear under both reporting bases and how competing interactions are handled. The table specifies evidence requests, not an assertion about every platform's implementation.

Read a vendor record within its limits

Kevel's Retail Media Guide, inspected on 6 September 2026, describes last-touch attribution involving impression or click events and purchase data. Its publication date was not established. This is a global first-party technical record, not an independently tested England deployment.

Use it to identify configuration questions for an actual Kevel implementation. Do not assume its impression terminology proves viewability under another standard. Obtain the precise adopted settings and precedence before treating its output as compatible with the comparison above.

A sales-credit total may change when rules change even if the underlying orders do not. Preserve the original configuration and revised definition so a reader can see whether the comparison concerns attribution settings or an observed commercial change.

Keep causal evaluation separate

The Magenta Book analytical methods, updated in May 2026, describe impact designs using treatment and comparison evidence. This UK government source informs evaluation reasoning; it does not turn click-based or view-based credit into a causal estimate.

For an incrementality question, commission a suitable design with its outcome, analysis unit, population, assumptions and uncertainty specified. Do not declare either attribution basis more causally accurate from an untested reporting comparison.

Record the unresolved choice

Select a reporting basis only after the actual evidence and intended reporting question are clear. There is no universal winner in this desk comparison. Record which reporting question the chosen basis is intended to answer. Publisher commercial relationships remain unverified; no paid inclusion or supplier recommendation is offered. Named independent and qualified measurement reviewers must check the final comparison before publication.