Understand content and paywall metrics
The Content Conversion table describes stored activity associated with content items. Its measures use recorded access decisions, rather than proof that an article was read or a wall appeared on screen.
Read the core columns
| Column | What it counts |
|---|---|
| Views | Recorded allowed and denied content-access outcomes for signed-in accounts and guest sessions, added together. Repeated requests can contribute more records. |
| Unique | Distinct identities counted separately within four groups: guest allowed, guest denied, signed-in allowed and signed-in denied. Those counts are added together for each article. |
| Granted % | Recorded allowed outcomes divided by Views, with the allowed count shown beside the percentage. |
| Paywalled % | Recorded denied outcomes divided by Views, with the denied count shown beside the percentage. |
Unique is not a count of people. One identity can contribute to both allowed and denied groups. Guest sessions and signed-in account IDs are different identities; changing session or signing in can add another contribution. The TOTAL summary adds the per-article Unique values, so the same account can also contribute across several articles.
The access-recording path requires a valid session and excludes requests identified as crawlers. Its allowed and denied results describe access decisions, rather than finished reading or a successfully displayed wall.
Work through one article
Suppose City Journal's aggregate for one article contains three recorded allowed outcomes and one recorded denied outcome:
| Measure | Result |
|---|---|
| Views | 4 recorded outcomes |
| Granted % | 75%, with a count of 3 |
| Paywalled % | 25%, with a count of 1 |
The example has four stored outcomes. A Unique count needs the identities behind those outcomes, which are not specified here.
The calculation rounds each fraction to two decimal places before the interface turns it into a whole percentage. Displayed percentages therefore have limited precision. When Views is zero, the displayed percentage is zero.
Keep the time boundary clear
Content Conversion reads stored aggregates from retained activity, without a selected activity-date window. Published filters content creation-date age, not the dates of the counted outcomes. These figures are not guaranteed to be complete lifetime histories.
Dashboard period cards and the reader record's activity summary use separate windows and counting rules. Start with Find content activity reports before comparing them.
Interpret the other columns cautiously
Signups % uses recorded signup actions associated with the content; Conversions % uses qualifying subscription-purchase transactions. Both percentages divide their counts by recorded paywall outcomes.
Conversions count distinct successful or free non-renewal subscription-purchase transactions for the same account and publication, strictly after a recorded denial and within 60 minutes. This associates the purchase with the content; it does not show that the article caused the purchase or received exclusive credit for it.
New MRR per view and New MRR per paywall are estimates derived from qualifying renewing subscription relationships. They are not cash collected. The TOTAL summary averages the per-content ratios, rather than establishing one publication-wide revenue-per-visit calculation.
Use Read payment reports for financial reporting. For a surprising figure, give the owner the exact column and report context using Investigate unexpected content figures.