The Revenue Model: The Shape of How Money Comes In

A revenue model is the arrangement by which money reaches a business: what is sold, to whom, on what trigger, and how often. Revenue has a shape as well as a size, so two businesses can report the same revenue and share nothing else. The shape is three things: when the money arrives against the work, how much of the cost moves with volume, and how much is known before the period starts.

Underneath that sits a reading habit almost everyone has, and it is worth breaking at the outset. A revenue figure appears, the eye reads it as a size, bigger beats smaller, and the comparison is finished inside a second. The size reading is not wrong. It is thin. A revenue figure is the sum of many separate arrivals, and the pattern of those arrivals carries information that the act of adding them up destroyed. Two lines can finish at exactly the same place and behave nothing alike the moment anything moves.

What is a revenue model, and why is it more than one number?

The same shape appears at a visible size on any street. Two food businesses stand on one lane. The first is a tea stall: it takes cash at the counter, roughly two hundred times a day, in amounts of ten and twenty rupees, and it knows almost nothing about tomorrow. The second is a caterer who books wedding work: it signs three orders in April for functions in November, takes a deposit on signing, does the work months later, and collects the balance afterwards. Suppose that at the end of the year both have taken the same amount of money. Nobody who has stood in either place would call them the same business.

A revenue figure is one number and a revenue model is at least three. The three are separable, and each one can be checked on its own. The first is timing: when the money is earned against when the work happens, and when cash actually moves. The three need not be the same instant. The second is volume sensitivity: how much of the cost attached to that revenue rises and falls as the quantity sold rises and falls. The third is advance knowledge: how much of next period's revenue is already committed before the period opens, and how much has to be won again from nothing.

Notice what the top lineRevenue: the first line of a statement of profit and loss. Everything else on the statement is subtracted below revenue, and that is why revenue is the top line. cannot show about any of the three. The total does not say whether the money came in evenly or in two frantic months, and an uneven arrival pattern of that kind is seasonalityA pattern that repeats at the same points of the calendar every year, such as heavier buying before a festival or at the start of a school session.. Nor does it say whether serving twice as many customers would cost twice as much. Nor does it say whether the same customers will be there in April. The three questions are the model, and none of their answers is visible in the total.

What is sold, and what makes the money count as earned?

Every revenue line has a trigger sitting under it: the specific event that turns an arrangement into revenue. Four triggers cover most of what an analyst will meet. A unit is shipped. A period elapses. A transaction completes. A right is granted for a stretch of time. Each one fires at a different moment relative to the work and relative to the cash, and once it is known which one a business runs on, a great deal about the shape of its revenue line follows without being told.

The trigger decides when revenue is recognised, and therefore what the line looks like. Consider two arrangements that raise identical money in a year. In the first, a marketplace takes a slice of every completed order, so its revenue line is a direct copy of buying behaviour: it bulges before a festival and thins in the monsoon. In the second, the same customers pay a fixed amount for a year of access, so the revenue line is twelve equal pieces regardless of what anybody does in October. Nothing about the customers changed. The trigger changed, and the line changed with it.

Two words are constantly used as if they meant the same thing, and they are worth keeping apart when reading a revenue line. BillingRaising an invoice and asking for money. Billing is an administrative act, and it can happen before, at, or after the moment revenue is treated as earned. is the act of asking for money. Recognition is the moment the money counts as earned. Where money arrives ahead of the trigger, it does not become revenue on arrival: it sits on the balance sheet as deferred incomeMoney already received for something not yet delivered. The business still owes the customer the goods or the service, so the amount is shown as a liability., a liability, and moves across into revenue only as the trigger fires. The gap between the two is why a business can collect a great deal of cash in a month and report very little revenue for it, and why the reverse also happens.

The work the business actually does between taking the order and firing the trigger is a separate study, covered under Operating Model and Supply Chain.

Which costs are the variable costs?

A cost is variable if it moves with volume. The test is worth applying in exactly those words rather than by feel. Feel produces a list of costs that seem small and flexible instead of a list of costs that actually track quantity. Ask the question about one extra unit: if Setu Bazaar, an invented online marketplace, serves one more buyer this year, does this cost go up? The payment charge on that buyer's orders goes up, so it is variable. The rent on the office does not, so it is not. The test says nothing about whether a cost is large, avoidable, or worth cutting. The test asks one thing, and the answer is yes or no.

Now the honest complication. Most first explanations leave it out. Most real costs are neither purely one nor the other. A delivery contract may carry a fixed monthly retainer plus a per parcel charge, so half of it answers yes and half answers no. Worse, a cost can answer no across the range under examination and yes across a wider one. A despatch centre costs the same whether it handles ten thousand parcels or its full capacity, so within that range it does not move at all. Push past capacity and a second centre is needed, and the cost jumps in one piece rather than rising smoothly. A cost shaped like that is flat inside a range and steps when the range is exceeded, and calling it fixed is only true as far as the edge of the range that happened to be examined.

Setting variable cost against fixed cost, and working out what that opposition does to a business as volume moves, is the subject of Fixed Costs vs Variable Costs: The Test Is Volume.

What is contribution margin, and what has been taken out to reach it?

Contribution is revenue less the costs that move with volume. Contribution margin is that same figure written as a share of revenue. Take Setu Bazaar's year: revenue of Rs 20,00,00,000/-, from which Rs 6,00,00,000/- is the cost of serving the orders and a further Rs 4,00,00,000/- is variable selling cost. Both answer yes to the volume test, so both come out. Rs 20,00,00,000/- less Rs 10,00,00,000/- leaves contribution of Rs 10,00,00,000/-, and against revenue of Rs 20,00,00,000/- that is a contribution margin of 50.00 per cent.

Check it a second way. A figure that only one route reaches is a figure nobody has checked. Setu Bazaar has 50,000 buyers and Rs 4,000/- of revenue from each. Rs 10,00,00,000/- of variable cost across 50,000 buyers is Rs 2,000/- a buyer, so each buyer leaves Rs 2,000/- behind after the costs that follow them, and Rs 2,000/- against Rs 4,000/- is 50.00 per cent again. The two routes are the same relationship rearranged rather than independent evidence, so they cannot disagree. The rearrangement shows that the margin is not an abstract ratio: it is Rs 2,000/- attached to one identifiable buyer, and it stays Rs 2,000/- whether there are fifty thousand buyers or five.

The intermediate step in that drawing, the Rs 14,00,00,000/- left after the first subtraction only, has a name of its own and a measure built on it. Comparing that measure with contribution margin, and being precise about which costs each one takes out, is the subject of Gross Margin vs Contribution Margin: What Each Subtracts.

Contribution margin answers what one more unit of revenue is worth to the business. The question is about the next unit and not about the average one, and the distinction does more work than it looks like it does. If Setu Bazaar wins one more buyer at Rs 4,000/-, the business does not keep Rs 4,000/-. Rs 2,000/- of cost walks in behind that buyer, so the business keeps Rs 2,000/-. A reader who takes the whole Rs 4,000/- as the gain from one more buyer has doubled the benefit of every growth plan in front of them. The contribution margin is the exchange rate between one more rupee of revenue and one more rupee available to cover everything that did not move.

Everything that did not move is still waiting underneath. Contribution set against costs that do not follow volume, and the count of units at which the two meet, is worked through in Unit Economics: Profitability at the Level of One Customer.

What is monetisation improvement, as against simply selling more?

There are two routes to a bigger revenue figure and they are not the same act. The first is more activity: more buyers, more orders, more units, the same charge applied to a larger base. The second is monetisation improvement. Monetisation improvement gets more revenue out of activity that was already happening, by changing what is charged, what is included in the charge, or which part of the transaction the charge attaches to. Setu Bazaar could take its slice on services it currently carries for nothing. Nobody buys more. The same flow of orders produces more revenue.

More activity and better monetisation are two different routes to the same larger number, and a reader who cannot tell which one produced a rise has learned nothing from the rise. The everyday version is a wedding caterer who books the same fourteen functions as last year but now charges separately for the serving staff who used to be included. Not one extra function was booked. The revenue has grown. The two rises came from different places and behave differently afterwards, so distinguishing that caterer from one who genuinely booked eighteen functions is not pedantry.

A third possibility hides between them and is worth naming so it does not get mistaken for either: a change of mixThe proportions of different products, services or customer types inside one total. Two periods can have the same total and completely different proportions inside it., where the same number of transactions at the same charges produces different revenue because the proportions inside the total shifted toward the more expensive items. Nothing was sold to more people and no charge was altered.

Weighing these routes against each other, and working out what each one costs to pursue, is what Revenue Growth vs Monetisation Improvement is for.

What does Setu Bazaar's revenue look like once it is given a shape?

Rs 500 crore of buying passes across Setu Bazaar in a year, it keeps 4.00 per cent of that flow, and so its revenue is Rs 20,00,00,000/-. Fifty thousand buyers produce that flow, and each one accounts for Rs 4,000/- of revenue. How a platform sets and keeps a slice of what passes through it was worked out in Take Rate: What a Platform Keeps of What Passes Through, and the figures here are the ones carried there. Each total below the revenue line can be rebuilt from its components rather than taken on trust.

Both columns add down as printed. Rs 20,00,00,000/- less Rs 6,00,00,000/- less Rs 4,00,00,000/- is Rs 10,00,00,000/-, a contribution margin of 50.00 per cent, and Rs 10,00,00,000/- less Rs 12,50,00,000/- leaves a result of minus Rs 2,50,00,000/- for the year. Per buyer the same arithmetic runs Rs 4,000/- less Rs 1,200/- less Rs 800/- to Rs 2,000/-, and Rs 2,000/- less Rs 2,500/- is minus Rs 500/-. Multiplying minus Rs 500/- by 50,000 buyers is back at minus Rs 2,50,00,000/-, and that check is worth doing every time.

Now hold the revenue line completely still and change only the shape. Suppose the same fifty thousand buyers paid Rs 4,000/- each as a yearly charge for access instead of paying it a slice at a time on transactions. Revenue is Rs 20,00,00,000/-, and the total is identical. The shape is not. When the money is earned, how evenly it arrives, and how much of next year is already spoken for all change. On a plausible retentionThe share of this year's customers who are still customers next year. A figure that has to be measured from actual behaviour, never assumed. assumption of 80.00 per cent, Rs 16,00,00,000/- of next year would be contracted before the year opened, against nothing at all under the transaction arrangement. The worth of that predictability, and its limits, belong to Recurring Revenue: Why Predictability Is Valued.

Why can two businesses at the same revenue behave nothing alike?

Set Setu Bazaar beside Anjani Stationers Private Limited, an invented notebook manufacturer. Anjani's revenue for the year is Rs 2,70,00,000/-, earned when goods are handed over and accepted by a dealer. Its published ladder runs rung by rung from the operating result down through finance costThe interest and related charges a business pays on money it has borrowed. Finance cost depends on how the business is funded rather than on how it trades, so it sits below the operating result. and tax, as follows. The rungs between revenue and the operating result were never published.

Read the ladder down and it adds as printed: Rs 41,50,000/- less finance cost of Rs 3,50,000/- is Rs 38,00,000/-, less tax of Rs 8,00,000/- is Rs 30,00,000/-, and with no dividend paid the whole Rs 30,00,000/- stays inside. The total is the one thing that survives being added up, so two businesses can carry the same revenue and share nothing else. Anjani Stationers earns on delivery, so its revenue line follows despatches and the school session behind them. Setu Bazaar earns on transactions, so its line follows festival buying. Neither pattern is visible in either revenue figure.

A shape shows what happens to this business if volume moves, and what part of next year is already settled. A shape is a description and not a verdict. A shape does not say whether the business earns. Earning needs everything below contribution as well. Whether the shape could be changed is a question about what customers would accept, and a shape does not answer that either. The worth of the business is a matter for valuation. Describing a revenue line completely and correctly leaves all three of those questions exactly where they were.

How does a reader recover a shape from a set of statements?

Three questions, asked in order, and the shape is drawn. A lender looking at a working capital limit asks them because the answers decide whether cash will be there in the months the repayment falls. An analyst building a picture of a business asks them because two lines that look alike in a spreadsheet will not behave alike. Anyone running a small business asks the same three about their own trading without calling them anything.

The three questions are when the money arrives against the work, how much of the cost moves with it, and how much of next period is already contracted. The first is answered from the trigger and from the gap between billing and collection. The second is answered by walking the cost lines and applying the volume test to each one. The third is answered from whatever is committed in advance: an order bookThe value of orders already received but not yet fulfilled. An order book shows what work is committed, though not whether the customer will still be there afterwards., a set of contracts with time left to run, or nothing at all. Three answers give the shape without a single extra figure.