A customer's worth across the whole life

Do not index

The Predictive Path

Course 2: Revenue as a system

Lesson 8: A customer's worth across the whole life

A customer's worth across the whole life

Why what a customer is worth is the net of its whole life, not the size of the deal that opened it.

The number you state most confidently about a customer — what they're worth — is the one you know least about at the moment you commit to it. Worth is the net of an entire relationship: acquisition, expansion, cost-to-serve, margin, the whole arc. It's committed early, when you can't yet see it, and resolved late, when it's too far along to change. This lesson is about reading a customer's economics across the whole life, and why what's still ahead matters more than what's already booked.

The number you trust most is the one you know least about.

The last lesson ended on a promise: that designing revenue depends on knowing, in real economic terms, what each segment and each customer is worth across its whole life. This lesson is about where that number comes from — and why the number most companies reach for in its place is the wrong one.

The substitute is the number struck at the sale. A deal closes. The contract value is booked. Someone divides what it cost to acquire the customer into the revenue it will bill and calls the result a payback period. A lifetime-value figure gets attached, usually by taking the contract and multiplying it out over an assumed number of years. These numbers travel: into the model, into the board deck, into the case for putting more money behind the channel that produced them. They feel like the economics of the customer.

They are struck at the single moment the company knows the least it will ever know about that customer.

At the sale, almost nothing has happened. The customer has not been onboarded. It has barely used the product. It has not renewed, expanded, or churned. No support load has landed. No margin has resolved. The only hard facts are the price agreed and the cost of winning it — and those two are the least predictive of everything the relationship will turn out to be. The company is pricing the customer at the opening line of the story and recording the figure as though the story were finished.

It is easy to watch this happen in a planning meeting. The channel with the lowest acquisition cost and the fastest payback is up on the screen, and the recommendation writes itself: put more behind it. No one in the room is wrong about the numbers on the slide — the numbers are real. They are simply the wrong numbers. They are the cost of getting the customers, and none of the cost, or the value, of keeping them.

So the most confident number in the model is also the least informed one. It is trusted exactly where it is weakest.

What a customer is worth is the net of its whole life.

The real worth of a customer is an economic arc, not an opening figure. It is everything the relationship produces and everything it consumes, from first contact to renewal or churn, netted out: the cost of acquiring it, the margin it throws off once it is running, the cost to serve it month after month, whether it expands or contracts, when and whether it leaves, and when the cash actually arrives. That whole arc is the customer's lifecycle unit economics. The deal size is one line inside it.

Laid out along the lifecycle, the pieces do not arrive together, and that is the heart of the problem. Acquisition cost is paid almost entirely up front, before the customer has produced anything at all. Revenue comes in instalments, across years. Cost to serve runs heaviest at the start, through onboarding, and then either settles or it does not. Expansion, when it comes, comes late. Margin — what is left after all of it — resolves last, because it is the residue of everything else.

So two customers that look identical the day they sign can resolve to entirely different worth once those arcs have run. The shape of the arc, not the size of the contract, is what decides which.

This is why lifetime value is not a figure you compute at the deal. It is a figure the lifecycle produces. The number struck at the sale is an input to it — the first input — not a summary of it. Treating that input as the answer is the mistake the rest of this lesson takes apart. And none of the arc is visible at the point most companies decide what a customer is worth. It becomes visible only across the path the customer actually travels.

The economics are committed at the sale and resolve long after it.

The structural reason the sale-day number misleads is a gap in time: the economics of a customer are committed early and resolve late, and the worth is decided in the space between.

Almost every figure that determines a customer's worth is set at or before the moment it enters. The segment it is drawn from sets the pattern it will most likely follow. The contract signed at the close sets the price, the term, the discount, the support tier. The onboarding path it is placed on sets how quickly it reaches value and how much it costs to get there. Course 2 has already called these the settings on the control board — and each setting commences a play that runs for months and carries its own economics. They are chosen in a moment. They pay out over years.

They also pay out in a fixed order. Cost to serve does not appear until the customer is being served. Expansion does not appear until the customer has had a reason to expand. Churn appears at the renewal, a year or more after the signature. Margin resolves last of all, because it is what remains once everything else has landed. This is the cascade the earlier lessons named, read in money: one commitment at entry releasing a chain of consequences that come to rest far downstream.

Which means the worth of a customer is not unknown at the sale because someone failed to measure it carefully enough. It is unknown because it has not happened yet. The relationship has been priced before the events that set its price have occurred. No sharper measurement at the moment of sale can recover the number, because the information is not there to be measured.

No single function holds the number.

There is a second reason the worth stays hidden, and it is structural rather than a matter of timing. Even once the arc has fully resolved, no one function is holding the whole number. Each holds a piece of it.

Course 1 named the four facts of a single customer — each decided at a different point, each owned by a different function, each carried in different terms. Those same four facts are the components of the customer's worth. Marketing holds the demand fact: which segment the customer came from, and what it cost to bring in. Sales holds the deal fact: the contract structure, the price, the discount, the support tier agreed at the close. Customer success holds the usage fact: how onboarding actually went, what support the account consumed, whether it adopted and expanded. Finance holds the margin fact: the cost to serve set against the revenue, resolved at the end.

Each of these is one term in the customer's lifecycle economics. None of them is the economics. The worth is the four of them netted together across the whole arc — and the four are never in one place at one time. They are produced at different stages, by different teams, in different units, recorded in different systems. The 360° picture the earlier lessons named — the customer's full record, assembled piece by piece along the lifecycle as each fact resolves — is the only place the worth actually lives.

This is why the number is absent from every dashboard. A dashboard belongs to a function, and a function holds a single fact. The worth sits across all of them, on the lifecycle, where no one function is looking.

Two customers, the same size, opposite worth.

Take two customers that close in the same quarter at the same contract value — sixty thousand a year each. On the sale-day numbers they are indistinguishable. They enter the model as the same customer, twice.

Then their lifecycles run.

The first onboards in a few weeks, mostly on its own. Its support load is light and stays light. It adopts steadily, and in the second year it grows into another twenty thousand of seats. It renews without a conversation. Across three years it bills well over two hundred thousand, costs little to serve, and resolves at a strong contribution margin. It turned out worth roughly what the sale-day figure implied — and then more.

The second needs a heavy, hands-on onboarding that drags on for months and never fully lands. Its support load runs well above the norm for an account its size. It never expands. At the first renewal, it churns. Across its whole short life it billed a single year of contract, consumed more service cost than a profitable customer consumes in three, and never repaid what it cost to win. Its lifecycle worth is not a smaller positive number. It is negative.

The same sixty thousand at the sale. One customer worth more than the figure said; the other worth less than nothing. The sale-day number could not tell them apart, because at the sale there was nothing yet to tell apart.

Now lift it to the segment. The segment with the best payback on paper — cheap to acquire, quick to close — can be the segment with the worst lifetime contribution, if the customers it brings in are mostly the second kind. And the deal the team celebrated at the close — the marquee logo won at a steep discount and a generous support tier — can be the account that runs at a loss for two years. None of that was visible in the number the company used to decide it was worth doing.

The number you spend against is what's left, not what's already done.

So far this lesson has treated worth as a verdict you read once a customer's life is over. But most of the decisions that depend on it are made while the customer is still in the middle of its life — part of the arc behind it, part still ahead. And for those decisions, the whole-life figure is not the number that matters.

At any point in the arc, the worth splits in two. There is the value already realized — revenue billed, cost incurred, margin booked. That part is done, and some of it is sunk. And there is the value still ahead — what the relationship is likely to produce and consume from here to the end. That is the remaining lifecycle value, and it is the only part any further decision can still change.

This is what makes lifecycle economics a working tool and not a scorecard. Every further investment a company can make in a customer — a marketing upsell campaign, a retention play, another round of hands-on support, a save on an account at risk — is exactly that, an investment, and the test is always the same: is the cost of the play justified by the remaining value it would protect or create? Not the value already banked, which the play cannot touch. Not the headline lifetime figure, which is mostly history by mid-life. The value still ahead, read from here. That is the denominator. Should we invest further in this customer? is the question lifecycle economics exists to answer — and it has to be answered again at every step, because the remaining value is different at every step.

Here the lesson rejoins the control board. Because the economics run the length of the lifecycle, the moment a play becomes worth running rarely lines up with its obvious trigger. A retention play is filed under the renewal — the contract date, month eleven or twelve. But the facts that decide that renewal resolved long before it: the onboarding that never fully landed, the support load that climbed in month six, the adoption that flattened in the first year. Read as remaining value, those facts already show the value ahead bending down — a year, sometimes two, before the renewal where the churn formally appears. The play the calendar files at month eleven is worth starting a year or more earlier, because that is when the remaining value first turned and the investment could still change the path. The dials set the arc in motion; reading the remaining value across it is what tells you the window opened early.

You still can't know it exactly — and acting as if you can is the error.

Reading the remaining value depends on knowing what customers like this one tend to have left at this point in the arc — and how well a company knows that depends on how much it has already seen.

For a brand-new segment — a market just entered, with no customers who have run their full course — the remaining value is a real unknown. The honest figure is a wide range, held with low confidence, and treated as the estimate it is. Pouring capital against it on the strength of an attractive payback is the error: it is spending real money against a number the company has not yet earned the right to trust.

For a segment with history, the picture is different. Once enough customers of a kind have run their lifecycles and resolved, what they had left at each stage stops being a guess and becomes a pattern. Course 2 has already named the shape that pattern takes: the lifecycle revenue curve, the way a segment's revenue, cost and margin tend to move over time. Read forward from a customer's current position, it is what tells you what this customer most likely has left — the remaining value the next decision is judged against.

The discipline, then, is to hold the uncertainty rather than paint over it. A remaining value read from many resolved lifecycles can be trusted and acted on. One assumed from a segment with no history is a guess, however precise the figure on the slide looks. Knowing which of the two you are holding is most of the skill.

What changes when the company reasons about worth this way.

When a company reasons about worth across the whole life instead of at the sale, the decisions capital rides on begin to change — at the portfolio level as well as the account level.

Acquisition spend stops being set against what a segment costs to acquire and starts being set against what it is worth once its customers resolve. A segment whose lifecycles produce durable margin earns more spend and more capacity, deliberately, because the company can see that it is worth it. A segment whose lifecycles resolve thin or negative has its inflow slowed before another year of cohorts is admitted on the same terms — not because a team failed, but because the worth, read across the life, said so.

This is also where the first course's most uncomfortable observation finally settles. Course 1 described margin that erodes with no visible mistake anywhere — the company growing, the reviews diligent, the number drifting down, the cause never quite locatable. Read through lifecycle economics, the cause is locatable. The company was scaling segments whose lifecycle worth was negative, with confidence, because the only number it ever looked at was the one struck at the sale, where every customer still looks the same. The erosion was not a failure of measurement. It was the predictable result of pricing customers before their economics existed and then allocating against the price.

All of this, though, is worth read from what has already happened — what is banked, and what the resolved facts say is still ahead. It can tell a company what a customer of a given kind has been worth, and what one in front of it most likely has left. The harder and more valuable move is to see the worth forming before the facts resolve — to read the path forward while the customer is early enough that more of it can still be changed. That is where the course turns next.

Next up

Once you can read the whole economic life forward, you can hold two of them side by side and choose.

→ Continue to Choosing between futures

———

This article is part of The Predictive Path

By Niko Laine, SaaS CFO

→ About the author

→ LinkedIn

You can read it on its own or explore the full curriculum.

→ View The Predictive Path

→ Explore courses