We’re living in a subscription world. More and more, everything that we do and use is no longer being sold to us directly. Instead, we’re now paying monthly for all kinds of things.
While a lot of press has been given to the Software-as-a-Service boom (SaaS), subscriptions are everywhere now. Some subscription models, like gym memberships and magazines, have been around for a long time, while other subscription businesses are brand new and innovative.
These days, we pay small monthly fees to access almost everything in our world. From music services, such as Spotify, Rdio, and Beats, to movie streaming sites like Netflix, to snacks and meal delivery like NatureBox and HelloFresh, it seems there’s a subscription service for anything and everything.
My local heating repair company is even getting in on the booming demand for subscription services and has launched a $10/month subscription service that includes basic system tune-ups, maintenance, and discounts on larger maintenance services.
Subscription services are popular because it often costs less, in the short run, to start using a service. For the price of buying one album on iTunes, I can get access to the entire world of music on Spotify for one month. For half the price of buying a new DVD, I can get access to the deep catalog of content available on Netflix. For a fraction of the cost of buying a treadmill, I can use an entire gym.
It’s no wonder that customers are flocking to subscription services to replace things that they once bought and owned outright. And, it’s no wonder that businesses are trying to come up with new and innovative subscription business models that they can offer to their customers.
If your business is building a subscription service, creating a reliable sales forecast is a critical step to understanding how your business will grow and what the key drivers of revenue growth will be.
Up next, I will walk you through the critical components of a subscription forecast, and show you exactly how to build your own. Our product, LivePlan, builds subscription forecasts automatically, so if you’re already a LivePlan user you’re in luck and can skip this article. But, if you’re using a spreadsheet and want to build your forecast yourself, you’ll want to follow along closely.
A few definitions you’ll need to know to build your subscription forecast:
When you’re building a subscription sales forecast, there are a few terms that you need to understand that you won’t find in your traditional sales forecast.
We’ll start simply with the Subscription Period, which is essentially the length of time that a customer commits to subscribing to your service. Many services operate on a monthly subscription period, meaning that customers are paying a monthly fee to access the service and can cancel at any time—think Netflix or your gym membership. Some services have annual subscription periods. In this case, customers are paying to get access to the service for a year, and can then renew at the end of the year. Magazines and enterprise software offerings often have annual subscription periods. In this article, I’ll be focusing on forecasting sales for a monthly subscription service. I’ll dig into annual subscriptions and other subscription lengths in a future post.
New Subscribers is also a fairly basic concept. This is simply the number of new customers that you sign up for your service in a given month.
Cancellations are exactly what it sounds like—these are the customers that choose to cancel your service each month.
Finally Churn, which is the ultimate number that any company with a subscription business model must pay very close attention to. Churn is the rate at which customers are canceling and leaving your service. Low churn equals happy customers, while high churn means that users are canceling quickly and not subscribing to your service for a very long time.
You calculate churn by taking the number of customers who cancel during a month and divide by the number of customers you had at the beginning of the month:
Churn = Cancellations During a Month ÷ Customers at Start of Month
But, there is a wrinkle to the concept of churn because it’s really just like a leaky bucket. Think of churn as a hole in the bottom of the bucket. As long as the hole (churn) is small, you can keep the bucket full simply by adding more water (new subscribers) than is draining out.
The definitions above are all you need to know to be able to forecast your customer growth. The next few things we’ll talk about relate to the revenue that you will make from these customers.
How to calculate revenue from subscription services
To start calculating how much money you will make, you need to estimate the Average Revenue per User/Customer (ARPU). This is how much you plan to charge a customer each month. If you are building a forecast that has different price tiers, you may want to create a different forecast for each price tier. We’ll talk more about this a bit later.
Using your churn percentage, you can easily calculate your Customer Lifetime Value (LTV). Your churn percentage helps you predict how long an average customer will subscribe to your service, and will help you predict how much money you expect to make from your customers over their projected lifetime with you. LTV is calculated by dividing Average Revenue per User by Churn:
LTV = ARPU ÷ Churn %
This calculation will tell you how much your average customer is worth to you. For example, if your service costs $19/month and your churn is 5 percent, then your average customer is worth $380 to you.
You can also calculate the expected customer lifetime of your average customer by dividing LTV by ARPU:
Customer Lifetime = LTV ÷ ARPU
Customer Lifetime can also be calculated by simply dividing 1 by your churn rate:
Customer Lifetime = 1 ÷ Churn
Using the same example from above, the average customer lifetime would be 20 months (380 ÷ 19 = 20) or (1 ÷ .05 = 20).
Finally, we’ll want to know how much money you forecast to make each month, otherwise known as Monthly Recurring Revenue (MRR). To calculate MRR, you just multiply your ARPU by the number of customers you have at the beginning of a given month plus the number of new customers you acquire:
MRR = ARPU x (Starting Subscribers + New Subscribers)
Now it’s time to build a forecast
Now that we have the basic terminology for subscription forecasting down, it’s time to look at building an actual forecast. As I mentioned, we’re going to focus in this article on building a simple forecast for a monthly subscription business, but many of the concepts discussed here can be used to build out a subscription forecast for different subscription periods.
Here’s what your spreadsheet will look like:
Again, this is a fairly simple model that assumes that your service doesn’t have any one-time setup fees and that your business operates a simple monthly subscription service where customers pre-pay for a month of service. The model can get more complex if users can upgrade their services over time or if you have one-time setup fees or different contract lengths.
In our sample model, you can adjust the Average Revenue per User (ARPU) each month if you expect that, on average, some customers might upgrade to higher-priced services that you offer.
Starting Subscribers is the number of customers that you have at the beginning of a month. In this forecast, we are starting with zero subscribers in our first month as a new business. In future months, we calculate starting subscribers by looking at how many customers we had at beginning of the previous month and then add in new subscribers and subtract customers that decided to cancel.
In this sample model, we assume that customers pre-pay for a month of service ahead of time, so we calculate churn rate based on the number of customers that we start each month with rather than the number of customers that we end the month with.
From there, we calculate churn, the projected lifetime of an average customer in months and in dollars and finally calculate the total monthly recurring revenue that you should expect to collect in a given month.
You can download this sample subscription sales forecast here and we will tackle more complex forecasting in a future post.
Questions? Please post them in the comments and I’ll do my best to help.