Measurement, control 12 of 17
M-12 Keep multimodal and Discover reporting distinct
Use the platform's actual surface definitions. Search generative impressions, Discover generative impressions and multimodal performance are not interchangeable.
- Priority
- High priority.
- Hard gate
- No
- Change class
- READ_ONLY: Inspect authorised evidence.
- Quality bucket
- Measurement and experiment readiness
Why it matters
Visibility from images and Lens behaves differently from standard organic results, and blending the two hides what is happening in each. For sites that rely on visuals, such as product, travel, local, food, property and visual how-to sites, this can be a large part of how people find them. Without a separate segment, changes in visual search go unnoticed.
Evidence to review
- Report manifests, available fields and a documented overlap policy.
How to implement
- Inspect which filters and reports exist for the property.
- Record each report's dimensions, attribution and time zone.
- Keep visual-heavy templates in separate cohorts when useful.
- Do not add report totals unless the documentation establishes they are disjoint.
How UltimateSEO checks this
Needs your evidence. No automated check covers this control, so it stays UNKNOWN until evidence is supplied.
Evidence needed: multimodal Search performance data from Search Console.
States used: PASS, FAIL, UNKNOWN, NOT_APPLICABLE or BLOCKED. UNKNOWN is never shown as a pass. How states work.
Sources
Part of The UltimateSEO Framework v4.3, catalogue dated 2026-10-14. Evidence states, control states and the publish gate are explained on the hub.