A subscription is at its core an arrangement with a customer that creates a periodic contractual commitment. Netflix is an example of a subscription-based business. The default in a subscription relationship is that the customer will make an additional purchase at some time in the future unless they take some action like sending a cancellation notice or not paying a renewal fee when the subscription ends. The other method used by businesses to sell products is a relationship that is based on transactions. Buying a hamburger at a Shake Shack is a transaction. Selling good in transactions does not mean that an ongoing customer relationship is not possible. Amazon sells products via subscriptions (e.g., Amazon Prime) and via transactions, but has an ongoing relationship with the customer in both cases. If a consumer buys a soft drink at a convenience store they do not have an ongoing relationship with that store, but they do if they use a loyalty card or use of a mobile app to pay at a retail establishment like Amazon Go.
A subscription-based approach to business is not a substitute for product market fit or a business model. Whether a subscription or transaction-based relationship creates the most value for a customer or a producer depends on many factors as will be explained below. The founders of a company like MoviePass can become confused about this reality because they see other businesses which appear to be successfully selling subscriptions. As another example of a mistaken understanding of subscriptions, the founders of startups may observe businesses like Dollar Shave Club (“Razors to Your Door for Just a Couple Dollars”) which was able to generate an attractive financial exit for shareholders and as a result the world ends up with thousands of business plans that are essentially “Dollar Shave Club for X.” Unfortunately for these founders, product market fit is still required even if a subscription approach is adopted. In addition, what has worked for other founders in the past may no longer work in the present.
Hosting a software driven service like Adobe Creative Cloud on servers instead of selling a version that runs on a local computer is natural to sell as a subscription, but the most important aspects of this model are things like creating a connected relationship with the customer, reducing uncertainty about whether the customer will pay for upgraded versions and being an effective way to reduce piracy. There are also less tangible aspects of a subscription relationship that that Robbie Baxter writes about in her book The Membership Economy, that I will not write about here. My focus is instead on the microeconomics and unit economics of the subscription relationship. Charlie Munger has said “microeconomics is what we do and macroeconomics is what we put up with.” As just one example of the business benefits of a subscription, in an essay entitled “Adobe’s Subscription Model & Why Platform Owners Should Care,” Ben Thompson writes: “to be relevant in a world where users interact with as many as five different devices in a day, then a per-device licensing model is clearly unsustainable.” After its shift to subscriptions Adobe no longer had to worry about things like when a customer would upgrade to a new version. Subscription-based businesses can have some very attractive attributes, but they also create new challenges which require new skills and approaches. Most importantly, as stated above, product market fit is still required.
Why don’t businesses sell subscriptions more often? After all, the revenue from a subscription is more predictable, the value of the data derived from the customer relationship can potentially be higher and other processing costs can be lower. What is not to like about subscription pricing? The answer is: it costs real money to get a customer to surrender the optionality associated with buying the product “just in time.” Buying products just in time increases the ability of the consumer to direct their cash to new uses if conditions change. Warren Buffett explains the value of cash to anyone including a consumer as follows: “cash is a call option with no expiration date, an option on every asset class, with no strike price.” Chris Rock puts it even more simply: “Wealth is not about having a lot of money; it’s about having a lot of options.” The less cash a consumer has available, the more valuable optionality is to that consumer. To illustrate this point, when a consumer is cash-poor they will do things like pay more on a unit basis just to buy just what they need in a “dollar store” or use a prepaid phone even though they pay more per unit of consumption because they want to preserve the optionality of what little cash they have.
Why do some businesses sell subscriptions but allow a customer to cancel at any time? Because the longer the customer is asked to commit to the periodic relationship, the more financial incentive the customers must be given to do so or the more in sales and marketing will be required to acquire the gross customer addition. In other words, the longer the commitment and the higher the commitment in terms of dollars, the higher the customer acquisition cost (CAC) will be. If this were not true, many people would be already paying more than $500 a month after making 30-year commitment to work out at their health club and businesses would routinely be making decades long commitments for software as a service. Some consumers are willing to pay more on a unit cost basis for a service that is terminable month-to-month than the would if they commit for a year or more for a reason. Would a business like a multi-year contractual commitment from its customers? Sure! But the CAC associated with that subscription offering may not create favorable unit economics for the business. It depends. Different prices may need to be charged to different customers as a result of the differing value placed on optionality.
When do consumers value a subscription enough that the “unit economics” associated with a subscription that I have written about many times before make sense for a business? I have a very smart friend who puts it this way: “If the value of the subscription is bigger than the price when compared to other alternatives, customers can be convinced to do it. It’s the same thing for an ice cream cone: if value minus price is bigger than other alternatives, eat up!” The value of any lost optionality is part of this equation. So that analysis naturally leads to an analysis of two different types of value associated with a subscription: (1) value for the consumer of the product versus buying in transactions and (2) value for the business providing the product in a subscription format. Every different product requires a unique analysis and that analysis must deal with a constantly shifting world that may present different and unpredictable alternatives and conditions. Consumers do this math in their head and are impacted by emotional factors in making decisions, as anyone who has seen a sales pitch for a time share condominium knows.
A subscription relationship with a customer has business attributes that resemble and annuity. The economics of a subscription business are very different than the economics of the factory that people read about in places like a college accounting class. To understand and operate a subscription business like this the managers and other employees of the business must acquire new skills and understand new concepts like lifetime value, cohort analysis and viral coefficients). These teams at a modern business using subscriptions must now include new types of employees like data scientists and use new technologies and approaches like machine learning. Since the business itself must make this changes and acquire these skills, investors must also be capable of understanding this phenomenon if they are going to be able to properly value the business. These changes I just described are very confusing for people who are unwilling to do more work than looking up a price/earnings ration on a web site. As I said, investors who are willing to do the work must learn new skills like calculating CAC payback periods. The unavoidable reality is that in valuing a subscription business, a NPV must be calculated for each customer and then the collective NPV of these customers must be aggregated. When confronted with the necessity of doing this work to understand a subscription based business model that resembles an annuity, many people say, “That’s too hard, let’s talk about P/E ratios.” That is a recipe for investing under-performance at best and financial disaster at worst. My blog post on how to value a subscription business is linked to in the End Notes. If you are unwilling to do this work, I suggest you buy a diversified portfolio of low cost index funds and spend more time on enjoyable hobbies.
McCarthy and Fader have written extensively on valuing subscription-based businesses:
The subscription business model is booming. Previously dominated by the likes of newspapers, magazines, gyms, utilities, and telecommunications firms, more products and services are being offered to more people through subscriptions than ever before. Business-to-consumer subscription businesses have attracted more than 11 million US subscribers in 2017, and the industry as a whole has been growing at 200% annually since 2011. There are over two thousand consumer-focused subscription businesses capitalizing upon customers’ diverse tastes. While many companies are selling more traditional products – such as food (Blue Apron and HelloFresh), grooming products (Dollar Shave Club and Harry’s), beauty supplies (Birchbox and Ipsy), and clothes (Stitch Fix and Trunk Club) While the industry is growing rapidly, it is also highly volatile.
Michael Mauboussin writes in his essay “The Economics of Customer Businesses”:
One of the keys to successfully analyzing a business is getting down to the most basic unit of economic analysis. For many consumer-oriented companies, and especially those that rely on a subscription model, the basic unit of analysis is the customer. Investors need to determine whether or not current and prospective customers add value. The only way to understand customer economics is to break down the analysis into its prime components. This customer-focused approach leads us to the following definition of value:
Value of a business = value per customer x number of current and future customers
In different words, business value is a function of customer economics and the current and future users (market size). This equation is simple, but unpacking the essential elements is not.
Fader and McCarthy describe a functional valuation process in their recent HBR article, in which they suggest:
…a “bottom-up” approach to corporate valuation …explicitly recognizing that every dollar of revenue that a company generates must come from a customer – and that not all customers are “created equal.” It is specifically suited for subscription businesses, driving revenue off of the flow of incoming customers over time, their retention patterns as they stay with (or abandon) their subscriptions, and the average revenue per customer (ARPU).
Let’s look at some of the factors that are potentially involved in a subscription-based business and who benefits from each attribute. In each case I suggest that you think about this key question: Does this aspect of the subscription relationship create producer or consumer benefits?
- “There is considerable evidence of consumer preferences for subscription over per-use pricing. There are three main reasons that probably lead consumers to prefer flat-rate pricing: (i) Insurance: It provides protection against sudden large bills. [Consumers Benefit] (ii) Overestimate of usage: Customers typically overestimate how much they use a service, with the ratio of their estimate to actual usage following a log-normal distribution. [Producers Benefit] (iii) Hassle factor: In a per-use situation, consumers keep worrying whether each [unit] is worth the money it costs. A flat-rate plan allows them not to worry [Consumers benefit.]” “[With subscriptions] the buyer faces a lower risk of price change: it is the seller that bears uncertainty about future prices. However, the payment in advance is a premium for running that risk.” [Consumers Benefit] Some products have more uncertainty than others, which can radically change perceived customer value. Subscribing to a health club is different than subscribing to a magazine. Both are different than buying a time share condominium in a tropical resort.
- Subscriptions can reduce monetary and other transaction costs between the seller and the buyer. [Producers and Consumers Benefit] The need for a transaction each time a unit of the product is delivered under the terms of a subscription can be reduced. For example, if a consumer owns a pet it may be simpler to agree to have food for the animal delivered on a regular basis. Someone who regularly consumes marijuana may want to receive a new bag of weed every week. Replenishment-based subscriptions have more relative value for consumers than deliveries of discretionary products like meal delivery and are not only less expensive to sell but experience less customer churn. Why? Replenishment of a supply like pet food is less of an optional purchase than a meal preparation kit which is more of a discretionary purchase.
- “In general, subscription plans also make it easier to develop close relations with customers. I wrote about this in my blog post entitled “Peloton: The “SaaS Plus a Box” Business case.” If access is on a strict per-use basis, there is less reason to obtain information about the users. But that does not stop vendors from trying to turn a transactional relationship with a customer into monetizable data. A subscription relationship lends itself more naturally to finding out what the consumers need, and to customization of offerings.” [Consumers and Producers Benefit]. A subscription relationship with a customer more naturally creates a valuable connected relationship with the end user. Better data capture via telemetry systems can lead to a more personalized offering for subscription-based firms than non-subscription-based firms (e.g., Stitch Fix versus Macy’s).
- If the price of a unit of consumption of a given product is lower than its price outside of the subscription this indicates that the subscription pricing model is being used by businesses to charge customers different prices for the same product or service (i.e., price discrimination). A classic case of price discrimination with subscriptions is magazines:
“consumers that assign a high evaluation to the magazine tend to purchase a subscription instead of purchasing the magazine at the newsstand; since the cover price is higher than the subscription price, the very same consumers pay a lower unitary price.”
As an example, The New Yorker magazine charges “12 weeks for $12” in its introductory subscription offer and one year for 89.99. The price for an individual copy at the news stand?
A car wash subscription is an interesting case since there is very little incremental cost of another wash just like a magazine. The Ben Thompson essay I noted above on subscriptions explores how subscriptions enable different customers to be charged different prices. [Producers and Consumers Benefit]
- When there is no subscription a business has to “win” a customer every time a purchase is made. When there is a subscription the default option is to stick with it even if it can be canceled at any time. This is an example of a pricing approach that benefits from a human tendency toward inertia. [Producers Benefit] Business love automatic subscriptions renewals precisely because of inertia and will often give customers a substantial benefit for agreeing to this arrangement. The reduction in churn can more than offset the cost of convincing the customers to accept the automatic renewal. Automatic renewals are so effective and potentially misleading that the FTC and numerous states have regulated the practice.
- Subscriptions and bundles are often coupled by businesses and this can create some confusion. One of the reasons why goods that can be naturally bundled are often sold with a subscription because a bundle is that they are likely to have something in them that is relevant for each time period. Spotify sells a bundle as does a university. Bundles are also more likely with information goods since they have little or no marginal cost and are nonrival and nonexcludable (public goods). In addition:
“There are reasons for producers, especially in areas like software, where network externalities are important, to also like these plans. Since per-use pricing does repress usage, it goes counter to the producer’s desire that a software package be used as much as possible in order to lock customers into that product. Producers would like consumers to become so used to the particular features and commands of their software that they will find it hard to change to another system. Producers also want their systems to be easy to try out, and be widely used, to capture additional customers. Subscription pricing and site licensing promote these goals.”
The best essay on bundles was written by Chris Dixon. There is no way to improve on it.
McCarthy, Fader and Hardie: https://lbsresearch.london.edu/586/1/Hardie_Valuing-subscription-based-businesses_mccarthy_et_al_16.pdf
My blog post on Amazon Prime and other Subscription Businesses: https://25iq.com/2017/07/15/amazon-prime-and-other-subscription-businesses-how-do-you-value-a-subscriber/
My blog post on Peloton: https://25iq.com/2018/03/10/peloton-the-saas-plus-a-box-business-case/
Aswath Damodaran http://pages.stern.nyu.edu/~adamodar/
Free versus Unit Pricing for Information Goods. http://firstmonday.org/ojs/index.php/fm/article/view/535/456
On the Economics of Subscriptions, Glazer, A. and Hassin, R. (1982), European Economic Review. https://www.sciencedirect.com/science/article/pii/S0014292182800597
Subscription as a Price Discrimination Device, Discussion Papers, Gabszewicz, J.J. and Sonnac, N. (1999), Recherches Economiques de Louvain, Italian print magazines and subscription discounts Andrea Mangani https://www.ec.unipi.it/documents/Ricerca/papers/2011-132.pdf
Adobe’s Subscription Model & Why Platform Owners Should Care, Ben Thompson, https://stratechery.com/2013/adobes-subscription-model-why-platform-owners-should-care/
The Great Unbundling, Ben Thompson, https://stratechery.com/2017/the-great-unbundling/
How Bundling Benefits Sellers and Buyers, Chris Dixon http://cdixon.org/2012/07/08/how-bundling-benefits-sellers-and-buyers/
The Economics of All-You-Can-Read E-Book Pricing: An Empirical Analysis http://www.fox.temple.edu/conferences/cist/papers/Session%203B/CIST_2015_3B_1.pdf
The Membership Economy: https://www.amazon.com/Membership-Economy-Forever-Transaction-Recurring/dp/0071839321