Pattern · Finance
Subscription and Recurring Revenue
Also known as Recurring revenue model, Servitisation, Retainer, Everything as a service. Anchored to Sandra Vandermerwe and Juan Rada named the shift from selling goods to selling ongoing service relationships; the modern subscription form has no single author., 1988.
Where this is contested
Subscription is older than any of the literature about it, since periodicals, insurance and utilities have all been sold this way for centuries. What changed in the late twentieth century was the deliberate conversion of product businesses into ongoing service relationships, which Vandermerwe and Rada named servitisation in 1988. The software-era vocabulary of churn, cohorts and lifetime value was assembled by practitioners in the 2000s and has no canonical text. The entry is anchored to the 1988 paper as the first serious treatment and carries the operational literature in the references.
Charging a regular fee for continuing access rather than a one-off price for a transaction, which converts selling from an event into a relationship and converts profit from something you earn once into something you have to keep deserving.
Plate · The shape
How it works
A transactional business earns when it sells. A subscription business earns while it keeps. That single change moves the centre of gravity of the whole company: the money is made after the sale rather than at it, so retention becomes the operating discipline and acquisition becomes an investment with a payback period.
The arithmetic is unforgiving in a specific way. You pay the acquisition cost up front and recover it over months or years, which means a growing subscription business consumes cash even while it looks increasingly profitable on paper. The faster it grows, the more cash it consumes. This is the single most common way an otherwise sound subscription business gets into trouble, and it surprises people every time because the profit and loss account is reassuring throughout.
The other thing that changes is what the business is for. In a transactional model the customer's experience after purchase is a cost. In a subscription model it is the entire revenue mechanism, because a customer who stops finding it worthwhile simply stops paying, and does so quietly, without complaint, on a date you will not notice until the month after.
The parts of the model
Continuing value
A reason to keep paying that renews itself. Either the thing is consumed and needs replenishing, or it improves, or it holds something the customer accumulates and would lose by leaving.
Signals
A customer who stopped paying would notice within a week · Value delivered in month twelve is at least as high as in month one · The reason to renew is not simply that cancelling is a nuisance
The commitment
Term, notice period, billing frequency and what happens at renewal. This is where most of the commercial argument lives and where most of the trust is won or lost.
Signals
The renewal is a decision the customer makes rather than one made for them · Cancellation is as easy as signing up, which is now a regulatory expectation as well as a decent one · Term length matches the payback period rather than being set by habit
Retention
The operational function of keeping people paying: onboarding, usage, support, relationship, and noticing when someone has quietly stopped engaging.
Signals
Someone owns the renewal number and it is not the salesperson who won the account · Declining usage triggers a conversation before the cancellation does · The business knows why people leave from asking, not from guessing
The cash gap
The distance between paying to acquire a customer and getting that money back. The financing requirement of the whole model and the number most often missed.
Signals
Acquisition cost and payback period are both known and tracked monthly · Growth plans are tested against cash rather than against profit · Annual up-front billing is priced deliberately rather than offered ad hoc
How you know you are in it
You do not fill a pattern in. You recognise yourself in it, or you do not. Read these as a list about your own business rather than as a definition.
- Revenue looks wonderful and the bank balance does not, and the faster you grow the wider that gap becomes.
- Your salespeople are paid for winning accounts and nobody is paid for keeping them.
- You can quote a churn figure and two people in the business calculate it differently, one by customers and one by revenue.
- Customers leave without complaining, and you find out at the point the direct debit fails rather than at the point they stopped caring.
- You have never separated new revenue from expansion revenue from churned revenue, so a flat month could be a good month and a bad month cancelling out.
- Somebody has suggested lengthening the contract term to fix retention, which is a way of not finding out why people leave.
The numbers that decide it
- Payback period on acquisition cost, in months. Under twelve is comfortable for most small businesses, and above twenty-four the business is a financing exercise whether or not anyone has said so.
- Churn measured two ways and reported as two numbers: customer churn and revenue churn. They diverge when you lose many small accounts or few large ones, and the divergence is the most useful diagnostic in the model.
- Net revenue retention, which is renewals plus expansion minus contraction and churn, on the existing base alone. Above one hundred per cent the business grows without selling anything new, which is the condition worth aiming at.
- Cohort retention curves rather than an average. Averages hide the shape, and the shape is what tells you whether you have an onboarding problem, a value problem or a pricing problem.
- Cash conversion under a growth plan. Model the next twelve months on cash, not on profit, and specifically model what happens if you double new sales, which for many subscription businesses produces a worse cash position rather than a better one.
When this shape works
- The need genuinely recurs, so the customer is buying something they would otherwise have to buy again anyway.
- The business can fund the cash gap, through up-front annual billing, a profitable transactional line alongside it, or capital that has been arranged deliberately rather than discovered to be necessary.
- Serving an existing customer costs less than winning a new one by a wide margin, which is what makes retention the cheapest growth available.
- There is something the customer accumulates by staying: data, history, configuration, standing, or simply a relationship with somebody who now knows their business.
And when it doesn't
- The need is genuinely occasional, in which case a subscription is a worse deal for the customer than paying when they need it, and they will eventually work that out.
- Value is delivered almost entirely at the start, as with an implementation or a one-off transformation, which makes month seven feel like paying for nothing.
- The business cannot fund the acquisition gap, which is the reason many owner-managed firms should convert gradually rather than wholesale.
- Retention would depend on the customer forgetting they are paying, which is both a poor business and, increasingly, a regulatory problem.
How this pattern dies
Growing into insolvency
The classic and the cruellest. Every new customer is profitable over their life and costs cash today, so a business that doubles its sales effort doubles its cash outflow while the profit and loss account improves. Firms have run out of money in the middle of their best year, and the warning signs were all in a cash flow forecast nobody was running.
The one-way conversion
A product business moves to subscription, revenue drops immediately because a one-off sale becomes twelve small payments, and the board loses its nerve in month five and reverses. The trough is arithmetic and entirely predictable, and it should be modelled and funded before the first customer is moved rather than explained afterwards.
Silent churn
Customers stop using the thing months before they stop paying for it, then cancel at renewal with no warning. The business had all the information it needed in its own usage data and was not looking at it. Retention that begins at the renewal conversation begins about four months too late.
Retention by friction
Keeping people through long notice periods, hard cancellation and automatic renewal. It works on the numbers for a while, it produces a customer base that resents you, and UK consumer protection rules have been moving steadily against it. A retention figure propped up by friction tells you nothing about whether the thing is any good.
Averaging the cohorts
Reporting a single churn percentage across a customer base with several quite different retention curves. The healthy segment subsidises the failing one in the average, and the business optimises for a customer it does not actually have.
In the wild
Rolls-Royce power by the hour
UK, and the reference case for servitisation. Airlines pay for engine flying hours rather than buying engines, which aligns the manufacturer's interest with reliability rather than with spare-part sales. The clearest demonstration that this pattern is about incentives as much as about billing.
The fixed-fee accountancy practice
Owner-managed, UK, and now widespread. Monthly retainers covering compliance, payroll and bookkeeping, replacing an annual bill that arrived after the work. It smoothed the practice's cash, smoothed the client's, and made the practice noticeably more saleable.
The maintenance contract
Owner-managed, UK, and the oldest small-business version of the pattern. Lifts, boilers, alarms, IT. It works when the equipment genuinely needs attention and fails when the contract is sold as insurance against a call-out that never comes, at which point the customer cancels the first time money is tight.
Adobe's move from licence to subscription
The best-documented conversion, including the revenue trough. Reported revenue fell for roughly two years as one-off licence sales became monthly payments, then recovered well past the previous level. Useful precisely because the painful middle is a matter of public record.
The owner-managed version
This is the pattern I would most often recommend to an owner-managed service business, with one loud caveat about cash.
The attraction is obvious and it is real. Recurring revenue smooths the year, makes planning possible, lets you hire before the work arrives rather than after, and materially improves what the business is worth if you ever sell it. A buyer will pay a different multiple for contracted recurring income than for a good reputation and a full diary, and the gap is not small.
The caveat is that converting costs cash before it produces any. If your annual bill becomes twelve monthly ones, you have just lent your entire client base eleven twelfths of a year's fees, and you have done it in the same month. Model that before you announce it. Convert in tranches rather than all at once, price annual payment up front at a discount that is genuinely attractive, and keep some transactional work running alongside while the base builds.
The part people underestimate is what it does to the business internally. In a transactional practice, once the job is done it is done. In a retainer practice, the client is judging you every month, quietly, and the judgement shows up as a cancellation eleven months later. Somebody has to own that. In a small firm it will be you at first, and the discipline is to notice the client who has gone quiet rather than the one who is complaining, because the one complaining is still engaged.
And do not lengthen the contract to fix retention. A twelve-month tie on a client who has stopped seeing the value buys you eleven months of resentment and a cancellation at the end of it. Find out why they went quiet instead.
Changing out of this shape
Convert in tranches, never wholesale. Model the cash trough first, including the month in which one-off revenue stops and monthly revenue has not yet accumulated, and arrange the funding for it before the first customer moves rather than in the middle. Start with the customers whose need most obviously recurs, since they are the ones for whom the change is a better deal and the conversion conversation is easy. Price annual up-front payment at a real discount, because that is the cheapest financing available to a small business and most owners leave it on the table. Instrument cohort retention from the first month, and give the renewal number an owner who is not the person who won the account. Moving back out of subscription is rare and usually signals that the need was occasional all along, in which case a credit-based or per-use model preserves the relationship without asking the customer to pay for months in which they wanted nothing.
Where to go from here
Frameworks that work inside this pattern
- Unit Economics
Tests whether a business makes money on each unit sold and each customer acquired, tracing revenue through variable costs to contribution and comparing the cost of acquiring a customer with the lifetime value they return. If the unit loses money, scale multiplies the loss.
- Cash Flow Forecasting
A rolling projection of cash in and cash out, opening balance to closing balance, that tells you when you're going to run short of money even while the profit and loss account still looks perfectly healthy.
- Value-Based Pricing
A pricing discipline that anchors price to the economic value an offer creates for a defined segment, with cost setting only the floor, and that tests willingness to pay before launch using tools such as Van Westendorp's Price Sensitivity Meter and Good-Better-Best tiering.
- Net Promoter Score
A customer loyalty metric built on one question, how likely are you to recommend us, scored 0 to 10. Subtract the percentage of detractors (0 to 6) from the percentage of promoters (9 and 10) for a single figure that is easy to track, easy to compare, and easy to abuse.
- Customer Journey Mapping
A visualisation of the end-to-end experience of a customer with an organisation, stage by stage and touchpoint by touchpoint, capturing what the customer does, thinks and feels, in order to find the moments that matter and the points where the experience breaks.
- Financial Ratio Analysis
Reads a set of accounts through the ratios that matter, liquidity, profitability, efficiency and gearing, to answer the two questions a business owner actually asks: can we afford this, and is the business actually healthy.
Playbooks that work the problem
- Pricing with Confidence
A sequence for putting a number on your work without flinching: find the job customers actually hire you for, price the value rather than the hours, check the volume the price must sustain, then imagine the new price list failed and find out why before it does.
- Turning a Business Round
Diagnosis before initiatives. Find out whether each sale makes money, where the losses pile up and what is throttling output, then point the whole organisation at the few fixes that matter. A week of honest arithmetic saves a year of energetic guesswork.
Neighbouring patterns
- The Productised Service
Bespoke work turned into something with a name, a fixed scope, a fixed price and a delivery system that does not depend on which clever person is free that week, so the business can be sold, staffed and improved rather than merely performed.
- Free and Freemium
Giving something substantial away at no charge so that somebody else pays for it: an advertiser, a subset of users who upgrade, or the same user later, which works when the free thing costs almost nothing to reproduce and fails quietly when it does not.
- Outcome and Value-Based Pricing
Setting the price from what the result is worth to the buyer rather than from what it costs to produce, and in its strongest form tying part of the fee to whether the result arrives, which pays very well when it works and moves real risk onto you when it does not.
What the critics say
The standard lifetime value formula, average revenue divided by churn rate, is wrong for most businesses. It assumes a constant retention probability across all customers and over time, when real cohorts show retention rates that rise as the weaker customers leave early. Applied naively it systematically undervalues long-lived customers and overvalues new ones, which distorts every acquisition decision made from it.
Fader, P. S. and Hardie, B. G. S. (2009) 'Probability Models for Customer-Base Analysis', Journal of Interactive Marketing, 23(1), pp. 61-69.
The retention economics that justify the model rest heavily on a 1990 claim that a five per cent improvement in retention raises profits by a quarter or more. The finding came from a small set of industry examples rather than from a controlled study, has been repeated for three decades with increasing confidence and diminishing sourcing, and should be treated as an illustration rather than a coefficient.
Reichheld, F. F. and Sasser, W. E. (1990) 'Zero Defections: Quality Comes to Services', Harvard Business Review, 68(5), pp. 105-111.
Servitisation frequently fails to pay. Studies of manufacturers moving to service-based models have repeatedly found a service paradox: substantial investment in service capability accompanied by lower returns, because the firm has taken on the cost and risk of ongoing delivery without the organisational capability to run it. Adding a subscription to a business that is structurally transactional generally produces this result.
Gebauer, H., Fleisch, E. and Friedli, T. (2005) 'Overcoming the Service Paradox in Manufacturing Companies', European Management Journal, 23(1), pp. 14-26.
Sources and further reading
- Vandermerwe, S. and Rada, J. (1988) 'Servitization of Business: Adding Value by Adding Services', European Management Journal, 6(4), pp. 314-324.
- Fader, P. S. and Hardie, B. G. S. (2009) 'Probability Models for Customer-Base Analysis', Journal of Interactive Marketing, 23(1), pp. 61-69.
- Gebauer, H., Fleisch, E. and Friedli, T. (2005) 'Overcoming the Service Paradox in Manufacturing Companies', European Management Journal, 23(1), pp. 14-26.
- Baines, T. S., Lightfoot, H. W., Benedettini, O. and Kay, J. M. (2009) 'The Servitization of Manufacturing: A Review of Literature and Reflection on Future Challenges', Journal of Manufacturing Technology Management, 20(5), pp. 547-567.