Skip to content
Edition 1 ReferenceDownloads Search Start here
Search
ChapterCH02 Chapter 2 of 12

Customer economics

What does the system optimise, and where does value appear?

The system seeks incremental contribution profit over a decision-appropriate horizon, subject to cash, customer experience, brand, capacity and strategic commitments. A short-term profit estimate cannot stand in for the whole strategy. Preserve future demand and obligations whose effects mature later.

Define the economic unit

Choose the entity before calculating: person, account, household, subscription or transaction. A reactivated subscription is not a newly acquired customer. Reconcile opening customers plus acquisitions and reactivations, less churn and other movements, to the closing stock. Do not multiply an already retained customer stock by retention again.

Record each metric’s numerator, denominator, entity, cohort, currency, period, cost scope, source, owner, freshness and version. Mark whether the estimate is observed, attributed, forecast or incremental. A change in definition is a version change, not an unexplained performance movement.

The economics tree

ModelM02Model
Incremental GTM contribution profit over horizon H
  = change in customer contribution versus the counterfactual
  − incremental intervention and acquisition expense not already included
  − implementation and operating expense not already included

Customer contribution
  = net realised revenue
  − agreed variable product, fulfilment, service and payment costs
  − returns, incentives and other agreed variable costs

Companion constraints
  cash and payback • future demand • customer harm • capacity • strategy

Use one cost classification throughout the calculation. If contribution already includes an incentive or service cost, do not subtract it again. A genuinely avoided expense increases profit once. Reallocated media is still expenditure if used elsewhere. Forecast future contribution and the same realised future margin are not two benefits.

Measure Meaning Decision limit
Blended CAC Agreed acquisition expense divided by all genuinely new customers Does not isolate the effect of paid activity
Attributed CAC Expense divided by customers credited by a rule or platform Is not incremental CAC; platforms can overlap
Incremental CAC Incremental acquisition expense divided by causally incremental customers Unstable or undefined when the effect denominator is small or spans zero
Marginal CAC Additional expense divided by additional incremental customers in a feasible spend interval Local to that tranche, population and period
Realised cohort contribution Observed contribution from an acquisition cohort through an agreed age Immature cohorts need explicit uncertainty
Contribution LTV Expected discounted future contribution over a stated horizon A forecast, not evidence that advertising caused the value
Payback Time until cumulative relevant contribution recovers acquisition investment State whether observed/forecast and discounted/undiscounted
Retention value Contribution with treatment minus contribution under the comparator, net of extra costs Never the whole LTV of everyone who stayed

Start with realised cohorts and cash payback. Add predicted lifetime contribution only when the forecast remains decision-useful out of time and across relevant cohorts. A short reliable horizon can be more useful than an elaborate distant prediction. For low-frequency purchases, use broad cohorts and conservative ranges.

Costs and customer scope must align. Use this model with the conditions and limits described above.

A consistent contribution LTV calculation

For each future period, multiply the probability of being active by expected contribution conditional on being active. Discount at a rate matching the period. Sum across the explicit horizon. Subtract the discounted expected post-acquisition costs only where they were excluded from that period contribution.

cLTV(H) = sum over t=1..H of
          [survival(t) × expected contribution(t | active)
           − expected additional cost(t)] / (1 + r)^t

The additional-cost term is unconditional expected cost for period t. Set it to zero when the contribution term already includes that cost. Acquisition cost stays outside this definition for the acquisition comparison. Use a model appropriate to the business, including nonrecurring purchases where a survival model does not fit. Editorial Resolution ER07: this makes the source’s time and cost convention explicit; it does not validate any forecast.

Compare the right alternatives

ModelM03Model
Option Description Benefit to isolate
A Current practice Baseline
B Repaired conventional process using existing/native tools B−A process gain
C B plus a bounded AI step C−B additional AI gain
D C plus extra coordination or autonomy D−C additional integration or autonomy gain

It is possible for C to beat A and still be a worse investment than B. Include review, rework, maintenance, integration, support, measurement, change and exit. Price five years for a durable commitment, using local assumptions rather than adopting the Red Team’s illustrative amounts. Show implementation cash, recurring cash and allocated staff opportunity cost separately.

Illustration, not a benchmark: a workflow saves 20 hours each week for 46 weeks. At CU60 an hour, that is CU55,200 of capacity. It is zero cash saving unless an expense disappears or an approved expense is avoided. If redeployment produces CU30,000 additional contribution, record that outcome; do not also book the CU55,200 as cash benefit.

Separate B−A, C−B and D−C. Use this model with the conditions and limits described above.

Make marginal thinking practical

GRW §6 illustrates a CU20,000 tranche that yields 43 incremental customers valued at CU600 contribution before media: CU25,800 less CU20,000 gives CU5,800 before any further costs. The next CU30,000 tranche produces 48 customers: CU28,800 less CU30,000 gives a CU1,200 loss. Historical average CAC can look acceptable while the next tranche destroys value.

Those customer counts and values are hypothetical. A real decision needs effect uncertainty, aligned horizons, cohort quality, cash, delivery capacity and alternatives. If incremental acquisitions plausibly include zero, present the customer-effect and profit range rather than a precise CAC ratio. “Hold the budget” is a valid decision.

Start here

Fill T2 for one cohort and reconcile it with Finance.

Then

compare the next feasible tranche with product, service and lifecycle alternatives.

Only when ready

use validated forecasts or response models.

Your next move
remove any benefit counted in more than one initiative.

Related

Evidence for this chapter6claims reviewed, with confidence ratings, limits and sources.

Claim classifications

Accepted means a principle or mechanism survived critical review, not that it guarantees a return. Accepted with conditions means prerequisites apply. Qualified means the original claim has been narrowed. Advanced needs substantial readiness. Experimental needs a capped comparison. Unresolved means the evidence cannot settle the claim. Rejected means the recommendation is excluded.

High, Medium and Low describe confidence in the stated target, not the chance of a positive ROI. For a Qualified or Rejected historical claim, confidence applies to the decision to narrow or reject it and to the supported limit—not to the original headline. Read the headline and classification together. An accounting identity or design principle is not empirical evidence of a commercial outcome. Follow the linked chapters, Plays and tools to see where each claim applies.