The one metric that actually correlates with revenue.
Completeness is the percentage of required attributes that have a non-empty value. That's it. The sophistication is in what counts as required — and for whom.
A flat completeness number is nearly useless. A per-channel, per-attribute-group completeness score is the single most powerful governance metric in product content, because every retailer onboarding gate and every ad network's content scoring rewards the same thing: complete, channel-shaped data.
Design the model, then the dashboard.
- Group attributes into regulated · logistics · marketing · media · optional. Never blend.
- For each group, set a minimum per channel — not one global target. The F&B and Retail playbooks spell out realistic numbers.
- Block publish when regulated-group completeness falls below threshold. Warn, don't block, for marketing.
- Expose the score in the authoring UI next to the SKU, not only in a weekly report nobody reads.
- Trend it. A flat 87% score means nothing; a falling trend is the alarm.
Targets that actually survive audits.
Completeness without relevance.
The single most common failure mode: a PIM dashboard showing 94% completeness while Amazon and Zalando keep rejecting listings.
Root cause: the 'required' set was defined by whoever owned the PIM, not by the channel. If your required-attribute set doesn't map to each channel's onboarding spec, a 100% score tells you your catalog is complete for nobody.
Fix: write the channel spec as a schema, import it into the PIM as a per-channel required set, and score against that. Every mature PIM supports this; most stock implementations skip it.