AVAIVUVPublisher Lab
Publisher metrics

Page RPM vs Impression RPM: Why the Denominator Matters

Both metrics normalize revenue to a thousand units, but they use different units. Mixing them can make a comparison look wrong even when the arithmetic is correct.

AIVUV Publisher LabSource-linked practical guide

Two ratios, two questions

Page RPM uses pageviews as its denominator. Impression RPM uses ad impressions. They can both be calculated from earnings, but they describe different layers of the publishing experience.

Page RPM = estimated earnings ÷ pageviews × 1,000
Impression RPM = estimated earnings ÷ ad impressions × 1,000

A worked example

Imagine a period with $200 in reported earnings, 40,000 pageviews and 80,000 ad impressions. Page RPM is $5.00. Impression RPM is $2.50. Neither is necessarily wrong: the first spreads earnings across views of pages, while the second spreads them across ad impressions. This example assumes both counts and earnings are from the same report window and compatible definitions.

Why the two values differ

A pageview is not the same event as an ad impression. A page can generate multiple ad impressions, no counted impressions, or different impression counts based on the page layout and the reporting product. The timing at which a system counts an ad impression can also matter. That means you should not compare one site’s page RPM with another report’s impression RPM and interpret the gap as a yield difference.

What to use for each question

  • Use page RPM to compare estimated earnings relative to pageview volume across comparable pages or periods.
  • Use impression RPM or the platform’s corresponding eCPM metric when the denominator is ad impressions.
  • Use total earnings to understand the reported monetary outcome.
  • Use both the ratio and its denominator to understand whether a change came with more or fewer opportunities.

Report labels matter

Reporting systems may offer separate metrics for ad requests, matched responses, rendered impressions or counted impressions. Each denominator describes a different stage in delivery. Read the exact definition shown by the platform before comparing the resulting rates.

Common mistakes

  1. Multiplying by 1,000 but dividing by the wrong volume.
  2. Combining earnings for one date range with views for another.
  3. Adding daily RPMs together instead of calculating the ratio from total earnings and total views.
  4. Treating the ratio as a prediction or as proof that one layout is better.

Calculate both from your report

Use the Page RPM Calculator for earnings and pageviews, or Impression eCPM Calculator for earnings and ad impressions. Use identical currencies, properties, date ranges and filters.

Why averaging rows can produce a misleading period RPM

Suppose one row represents 1,000 pageviews and $10 earnings, while a second row represents 10,000 pageviews and $20. The first row has a $10 RPM; the second has a $2 RPM. The average of the two row RPMs is $6, but period RPM from totals is $30 ÷ 11,000 × 1,000, or about $2.73. For a period-level RPM, calculate from summed earnings and summed denominator. A simple average of row ratios answers a different question and can overweight low-volume groups.

Check the metric name in exports

One report may label a metric “Ad RPM,” another “Impression RPM,” and another “eCPM.” The calculation can be similar while the exact impression stage or revenue definition differs. Follow the definition in the source product and avoid mixing metrics from separate platforms without understanding their denominators.

For practical reporting, store the original metric label with the exported data, then calculate a comparable ratio from the corresponding totals. A number without its denominator and time range is not a reliable comparison.

Sources

Continue with the tool: Open the related AIVUV tool →