The benefits of subscription-based services are by now very well-known, and while in the past we had to convince startups to shift to a recurring revenue model, almost all startups these days try to implement one. Not only is the recurring revenue business model much more manageable than charging a one-time fee for providing a service, but these companies also tend to be valued on a higher revenue multiple than companies who continue to sell one-time (perpetual) licenses.
A subscription model is very natural for products which are used on an ongoing basis, such as a sales tool or an HR software license. But on the other hand, there is a growing number of startups which provide services that are much less recurring in nature, yet are ‘forcing’ their customers to purchase a subscription instead of paying for a single use. The general logic behind pushing for subscription is that the company can report revenue as Annual Recurring Revenue (ARR) and therefore benefit from ARR multiples when raising capital.
Let’s take, for example, a startup that provides a new way to do business card design. Naturally, a business card is not changed frequently and therefore is likely not a good fit for a subscription model. But let’s assume that instead of charging $50 for a single design, the startup now charges the same $50 as an annual subscription, allowing the customer to make several edits throughout the year. Most likely, many customers won’t need another business card after a year and will not renew the service. Some customers such as design studios may renew the annual license while others will just forget to cancel it. However, the result will be high customer churn after the first year.
Let’s assume 30% of the customers who signed up for such a service renew after a year, and for simplicity, let’s assume 90% of the customers who stay after a year keep renewing the service every following year. The retention cohort will look something like this:
What this implies is that within a year the company will lose 70% of the revenue it generates from customers who signed up for the service today. If the company plans to double revenue next year, it will have to first fill in that 70% of lost revenue from its existing customer base and then grow another 100%. So for it to grow 2x, it will need to generate 170% of new revenue instead of 100%. But this is still misleading — as the 170% of new revenue will also suffer from the same high churn and will result in much lower revenue down the line. I am not going to go over the math, but the ‘real’ new revenue number the company needs to attain in order to double revenue while taking into account the large first-year churn is 566%!! Put differently, this means that instead of adding 100% of new revenue every year the startup will need to add 566% to achieve the same growth rate. This is similar to trying to fill a leaking bathtub with a bucket that has a huge hole in the bottom. It is a daunting effort.
Now let’s assume this startup doubles ARR every year for the first five years and then stops adding new customers. As you can see in the chart below, ‘real’ ARR in year six is only 42% of what was reported a year before:
What this means is that at some point in the future high-churn businesses will likely suffer from a stark decline in growth which can have a terrible effect on the business. It is true that, for most companies, growth typically declines over time as revenue scales, but this decline is much more gradual for “regular” businesses and can be accounted for in their budgeting and plans. But companies who suffer from high churn witness a very rapid decline in growth once they stop feeding the Sales and Marketing beast.
Many consumer and SMB subscription companies suffer from low first year retention, and it doesn’t necessarily mean these aren’t good companies. But for the above reasons, high-churn businesses should not be valued on the same revenue multiple that a ‘typical’ SaaS businesses attracts. These business models are just much more difficult to sustain over time as the bathtub keeps on leaking, and the hole in the bucket just grows and grows every year.