Why successful publishers focus on Session RPM

The amount of data available to a web publisher can be a two-edged sword. On one hand, more data means being able to inform a wider range of decisions. On the other hand, that data can distract from what is important or even overwhelm. A common trait amongst successful publishers is having focus on the metrics that matter most. At OKO we focus on Session RPM as our headline metric. In this blog I’ll argue the case for that metric and look at the reasons why we, and other successful publishers look at Session RPM over other metrics. By the end I hope that you will be on board with the value of Session RPM and be better equipped to increase your website revenue.

What is Session RPM

Session RPM is simply your earnings per thousand visits to your website. The RPM part refers to “Revenue Per Mille”, or revenue per thousand. The formula  is simple: Revenue / Sessions * 1000.

To determine an accurate Session RPM you need to ensure that it is calculated with Total revenue from all sources and all sessions.

Why is Session RPM such a powerful metric?

Session RPM brings the focus on to what you actually earn per visit. It removes the impact of traffic fluctuations and accounts for the impact of user experience & flow, complex ad set-ups and common issues to give publishers a true yardstick to measure revenue performance. To really understand the benefits of thinking in terms of Session RPM, we’ll look at examples of how it is more meaningful than other popular metrics:

Session RPM vs Total Revenue

Total revenue is a popular metric for obvious reasons: It is the only one that you can actually spend. It’s an important metric to track, but allows traffic fluctuations to hide changes in revenue performance.

Example: A new ad unit is introduced in the same week as a significant drop in traffic. Total revenue might be down due to the traffic change despite the new unit bringing additional revenue. Session RPM would show the gains from the new unit and allow the most revenue to be gained from the traffic that the site does have.

Session RPM vs eCPM (Effective CPM)

eCPM is calculated from revenue and ad requests, meaning that it factors in fill rates. As such it is a useful metric for making trafficking decisions, but a lousy one of revenue optimization as it fails to see beyond the individual ad request.

Example: A page has a single ad unit earning an eCPM of $3.00. A second unit is added. This doesn’t impact the original, but earns less at an eCPM of $1.00. Looking at eCPM would show a 50% drop to $2.00 despite this potentially adding 33% to earnings.

Session RPM vs Page RPM

AdSense in particular like to use Page RPM as a metric, which is a is a great improvement over metrics that only measure isolated ad units. AdSense’s Page RPM figure only has value if all ads on the site are served by AdSense, but Page RPM in general still fails to take into account how bad User Experience can decrease revenue.

Example: A website averages $4 Page RPM with users viewing an average of 3 pages per visit. Additional ad units are added to key landing pages, bringing the Page RPM up to $6. This seemingly positive result hides the fact that pageviews per visit drops to 1.5. Session RPM shows the real picture that the revenue per 1000 visits has dropped from 25% from $12 to $9, which would impact revenue by a similar proportion.

Session RPM vs CPM (Impression CPM)

Impression CPM is one of the most used metrics by publishers, despite being a metric for buyers that has very limited use for publishers. Impression CPM is simply the amount paid for 1000 impressions (in publisher terms this is often net of fees and rev share). Although it seems logical that buyers paying more means higher earnings, there are numerous situations when this isn’t the case. Let’s look at one simple example:

Example: A publisher averages $3 Impression CPM. A new partner offers $5, which seems like a good deal. Unfortunately the new partner is only fills 50% of impressions, so they actually earn less per potential impression (ad request).

Session RPM vs Ad Session RPM

AdSense introduced Ad Session RPM  as a way to tackle the issues in this article. For publishers using only AdSense it represents a  big step forward, but still has two significant issues. First is simply that it only takes AdSense into account. Swap out an AdSense unit for a higher paying one from another partner and the AdSession RPM will drop despite this being a wise move. Even for AdSense only publishers, there is an issue around the measurement being based on ad sessions rather than true sessions.

Example: A website with an Ad Session RPM of $12 has a sudden increase in adblocker usage by its users. Ad Session RPM actually increases (as many of these users were already less valuable to advertisers), despite this negatively impacting revenue. True Session RPM would reflect this change.

If Session RPM is so good, why isn’t it used more?

There is one simple reason that Session RPM isn’t used as widely as these other metrics and that is that it is a more complicated metric to measure. To get an accurate Session RPM means totalling revenue from all sources and dividing that by Sessions from a platform like Google Analytics. If you work with multiple ad partners (which most publishers should), no single partner can tell you the Session RPM, which means it needs to be worked out. If done manually this usually means a weekly or monthly “spot check” in a spreadsheet. Not a hard job, but one that many publishers overlook.

OKO publishers thankfully don’t need to do that manual checking. Our system automatically pulls revenue from all managed sources and compares this to Analytics traffic for accurate daily figures. However, even if doing the calculations manually, they are well worth doing.

Summing up: The case for Session RPM

Session RPM is the simplest, most meaningful metric that actually allows publishers to measure the revenue impact of changes. Because it combines two absolute values (revenue earned and visits), Session RPM rolls the impact of user experience, placement, trafficking, latency, dropped impressions, actual rates and more into a simple, easy to track metric. As such, it is no wonder that is the favourite metric of the OKO team and the majority of the most successful publishers that we work with.