Revenue Retention Analysis: Interpretation

June 16, 2010

This blog post is about interpreting data in a billings/revenue retention analysis, and part of a series of posts that serve as a step-by-step guide on conducting the analysis from start to finish. Beyond the insights you will gain, conducting the analysis will be helpful for most expansion stage companies hoping to raise expansion capital. Many venture capital firms will perform this analysis at some point during the due diligence process. Presenting this data upfront will save them time and likely impress their management teams with your “metrics-driven approach” to management.

In my last post, we examined and interpreted a chart that illustrated total billings over time by cohort (shown below).
To recap, the chart shows that, in general, billings increase over time, and that most of the cohorts act like a perpetuity with growth. On top of that, subsequent cohorts started at higher billings amounts than preceding cohorts, indicating that the company is signing up more customers each quarter, signing up larger customers, or both. In general, the output of this analysis is very desirable, because the effect of upsells are more than the counterbalancing churn.

While billings for most of the cohorts are increasing over time, there were two anomalies: the Q3 and Q4 2008 cohorts (whose total billings decreased dramatically in month 2 and then started increasing gradually after that). I speculated that the underlying cause of the month 2 decline in these cohorts could have been a product issue, the company signing up customers who did not find the product useful, or perhaps that the company billed some of its customers in Q3 and Q4 upfront for a year, which resulted in inflated billings in month 1, and much lower billings in months 2-12.

To understand if the latter was the cause of the month 2 decline, we must examine the average customer bill by cohort. If the average bill declined proportionally in month 2, it would indicate that customers are not churning and that upfront billing is the probable cause of this anomaly. A chart that illustrates the average customer bill by cohort over time is shown below:

From this chart, we can tell that the decline in the average customer bill from month 1 to month 2 is similar to the decrease in total billings (about a 75% decrease). This indicates that it is indeed likely that the company billed some Q3 and Q4 2008 customers upfront, which was the cause of the month 2 decline. A closer examination of the data set confirms that a number of customers were billed upfront for a year, and after the year was up, were billed on a month-to-month basis.

The moral of this story is that interpreting the results of a revenue retention analysis can be tricky, and that it is easy to jump to a specious conclusion that provides a plausible explanation for something such as the Q3 and Q4 anomalies. When interpreting a revenue retention analysis, it is important to understand that the spend behavior of customers is only a symptom of potential issues/opportunities, and usually, there are multiple possible explanations for any trend or anomaly. A revenue retention analysis uncovers symptoms of problems, but not always their causes. So be careful when interpreting the data.

Beyond that, the average customer bill chart shows that customers pay $300-$400 on average per month, and that this amount increases over time. Further, there is an upward trend in the average customer bill by cohort as most of the cohorts start at a higher average bill than a previous cohort.

CEO

Vlad is a CEO at <a href="http://www.scan-dent.com">Scandent</a>, which develops radio frequency identification (RFID) systems that prevent theft, loss, and wandering/elopement in hospitals and nursing facilities. Previously, he was an Associate at OpenView.