The new version of CPI detects anomalies in the metrics and acts on them by lowering the importance of these metrics in calculations. As a result, we have an algorithm that prevents one metric from going wild and taking the whole CPI with it. Also, articles that don't fit in the general pattern have a better chance of scoring higher values than before.
Understanding the new Exposure CPI score
Let's see how the new calculations affect the Exposure CPI score using the example of an article with a high number of social actions, but lower than expected visits on the site.
The second version of the algorithm would allow the high number of social actions to push Exposure CPI higher than deserved.
But, with its third, perfected version, the algorithm will know that behavior on social networks is not in-tune with the behavior of readers on the site. It is smart enough to adjust the importance of the social actions for that article when calculating Exposure CPI.
Understanding the new Engagement CPI score
As for the Engagement CPI score, the algorithm is smart enough to know which articles are different from the others published on the same website, and what behavior could be considered as expected.
For example, in the case of the longer articles, the algorithm knows that it cannot expect the same values of Read Depth as it can with shorter articles. This allows a fairer treatment of longer pieces when it comes to engagement.
The new version of the CPI values harmony among metrics - even when all numbers paint the same picture about reader behavior.
This is accomplished by taking into account the weight of metrics, which are upgraded and calculated on the fly by using non-linear functions. Functions establish what "normal" is and what should be considered as an "anomaly" during each calculation.
New Loyalty CPI is a revolution for itself.
Up until now, we have been able to measure how many articles managed to attract highly engaged returning users.
But with the new Loyalty CPI, it is now possible to measure an article's ability to engage Loyal Readers.
Loyalty CPI doesn't work on aggregated data anymore. The third - and most advanced - version of the algorithm, first establishes two things:
- Loyal reader base
- The degree of loyalty of each reader on the site
Then, it looks at the volume, percentage, and quality of those loyal readers in the readership of the observed article. These three factors are used to calculate the Loyalty CPI of each individual article.
Why does the Loyalty CPI matter for your publication?
With this new approach in calculations, you can finally answer the questions such as:
- Which content is most loved by your loyal readers?
- Where should you look for patterns that nurture loyalty among our readers?
- Which content is driving subscriptions?
The answer to all these questions is: high Loyalty CPI score. It can be safely used as your "north star" metric in this context.
How does the industry define a "loyal reader"?
Up until now, we defined a loyal reader as "highly engaged returning reader". Despite the fact that it was the most advanced definition of reader loyalty commonly used in the industry, we saw that there were a few issues with this definition:
1. "Loyalty" is a complex notion, tied to one person's behavior, so it must be measured on the visitor level. In the past, our infrastructure allowed measurement only on an aggregated level.
2. The definition of the "returning reader" is not really informative. The fact that somebody returned to a website doesn't really tell us anything about the reader's relationship with the publication in question. And it most certainly doesn't tell us how loyal the reader is.
While waiting for our infrastructure to catch up and allow us to measure individual readers (without any personal data, fully GDPR compliant), our Data Science team experimented with a lot of things. Even giving more value to advanced metrics like recency and frequency was simply not enough.
How does Content Insights define a "loyal reader"?
We needed something that could tell us if your site is truly a part of a visitor's routine: if they come to your site regularly, day by day.
So, we started to measure "Active Days" of each reader - a day is when they opened at least one article on your website. When readers have more active than inactive days, we call them "sequential".
Or, to be even more precise, the data has to show they have built a habit around your publication and that it is indeed a part of their reading routine. You can think of it as "returning" but with enough depth and certainty for the notion to actually have value for editors.
This is why we expanded our old definition:
A Loyal reader is a habitually highly engaged reader. It's a visitor who reads articles on your site as part of their routine, and finds them engaging most of the time.
Only when you are part of a person's routine and they like it that way - can you say that you have a meaningful relationship with them that you can call loyalty. And the degree of engagement of this particular reader base is actually a degree of their loyalty.
By measuring things this way, we are actually measuring the momentum of their relationship with your site like never before. We can detect when that momentum is on the rise, and we can detect when a loyal reader is slipping away.
Knowing this is crucial for your editorial strategy, especially if you're operating on a subscription business model. You will know which content is valued more by your loyal readers and thanks to these insights - you will be able to develop meaningful strategies and make them stick around.