OPERATING BRIEF / 01

Unit economics is the instrument panel for growth.

A top line is an output. It does not tell you whether the next unit creates cash, consumes cash, or requires an invisible subsidy from the founder, the support team, or the balance sheet.

Define the unit, isolate the variable work, and trace the customer from acquisition to collection. The weekly dashboard is the shortest path between an economic assumption and the evidence that can confirm or kill it.

CONTROL SEQUENCE

Four instruments. One system of record.

Use the sequence to inspect contribution, recovery, retention, and cash in order. Each node contains the calculation and the next evidence request.

CONTROL 01 / MARGIN

Find the contribution leak

Start with realized price and every variable cost that moves with the unit. A growing top line can still be a warning when delivery, support, usage, refunds, or payment costs consume the contribution left after acquisition.

realized price − variable cost = contribution
SCHEMATIC / UNIT CONTROL

Do not collapse four different questions into LTV:CAC.

LTV:CAC is a model output. The dashboard below exposes the inputs that move first and the operating decisions attached to them.

MOD_01

Contribution margin

The unit is the smallest repeatable economic object in the business. It may be an order, account, project, location, or month of service. The definition must match the way costs scale. When service hours, support, onboarding, infrastructure, payment processing, refunds, or fulfillment rise with volume, the variable-cost ledger must show them.

CONTRIBUTION / UNITrealized revenue − variable costShow both dollars and percentage. Keep list price, collected price, and concessions separate.

Probe: did realized price fall, did variable work rise, or did the mix shift toward a unit that was never profitable?

MOD_02

CAC payback

Acquisition is an investment. Payback asks how long the new customer must generate contribution before the investment is recovered. Use a blended definition for the company view and a channel definition when the operator is deciding where the next dollar should go.

PAYBACK / MONTHSCAC ÷ monthly contributionUse contribution, not revenue. Separate contractual timing from collected-cash timing.

Probe: is the payback extending because CAC is higher, contribution is lower, sales cycles are slower, or first-90-day service cost is being deferred?

MOD_03

Cohort retention

Logo churn and revenue churn answer different questions. Gross retention shows what remains before expansion is counted. Net retention adds expansion and reactivation. Cohorts protect the analysis from the false comfort of a young customer base or a few large expansions.

GRR / PERIOD(beginning MRR − churn − contraction) ÷ beginning MRRPair with activation, usage, support intensity, and realized contribution.

Probe: is the newest cohort reaching first value, renewing, and expanding at the same rate as earlier cohorts?

MOD_04

Cash efficiency

Growth quality is ultimately tested by cash. Reconcile contracted revenue with invoices, collections, failed payments, refunds, and the variable costs paid before customer cash arrived. The rolling view is usually more useful than a noisy single week.

BURN MULTIPLEnet burn ÷ net new ARRState the burn definition and use the same window for numerator and denominator.

Probe: did the business buy growth by financing receivables, delivery, inventory, or acquisition before the unit paid back?

DATA LEDGER

One weekly table should explain the movement.

Build the ledger from source events. Ratios are the final layer, not the raw material.

LayerMeasureFormulaDecision
PriceRealized revenue / unitCollected revenue ÷ unitsInvestigate discount, credits, refunds, mix.
DeliveryPost-service contributionRevenue − COGS − acquisition-variable − service-variableReprice, redesign, narrow, or remove the unit.
AcquisitionCAC paybackCAC ÷ monthly contributionChange channel, offer, sales process, or terms.
RetentionGRR / NRRBeginning revenue adjusted for movementsTrace cohort, activation, usage, and expansion.
CashBurn multipleNet burn ÷ net new ARRSlow spend, improve collections, or fix contribution.

Separate actuals from assumptions. A modeled lifetime, forecasted expansion rate, or expected margin belongs in a planning layer. The weekly control layer should show what happened to the actual cohort: what was collected, what was delivered, what it cost, and whether the customer is still producing contribution.

Segment before averaging. A blended metric can be directionally useful and operationally useless. Split by channel, offer, plan, customer size, geography, acquisition month, and delivery model when those dimensions can change economics. The question is not “what is our CAC?” It is “which units are we buying, at what cost, with what future burden?”

FAULT ISOLATION

Margin death leaves a trace before it leaves a hole.

Use linked variance patterns to select the next diagnostic test.

TRACE_A

Revenue rises / contribution falls

Pull realized price, refund, credit, service-hour, usage, and support data. The likely failure is a unit that looks larger on the invoice than it is in delivery.

TRACE_B

CAC rises / close rate holds

Pull spend allocation, sales labor, sales-cycle length, deal size, discount rate, and channel mix. The channel may be buying harder-to-serve customers.

TRACE_C

NRR holds / GRR falls

Pull account-level movement and concentration. Expansion can hide contraction when a small number of customers upgrade while the base leaks.

TRACE_D

ARR rises / cash falls

Pull collections, aged receivables, payment failures, delivery timing, and burn. Contract value is not liquidity until it is collected.

Alert rule: investigate when two linked measures worsen for two consecutive periods, or when one movement threatens cash or delivery capacity. Use thresholds as internal controls, not universal laws.
FIELD QUESTIONS

Questions for the weekly operator review.

Ask the question, name the evidence, assign the owner, and record the next reversible action.

Why not use a 3:1 LTV:CAC target as the dashboard?

Because LTV is forecast-driven and can remain attractive while current contribution, payback, retention, or cash timing deteriorate. A composite ratio is useful only when its components are visible and reconciled to actual cohorts.

Which costs belong in contribution margin?

Include costs that genuinely move with the unit: direct production, fulfillment, payment fees, usage infrastructure, delivery labor, onboarding labor, and incremental support. Keep the list stable and disclose costs that are partly variable.

How should a small company handle noisy weekly ratios?

Keep the weekly close, but use rolling four-week or thirteen-week views and inspect the underlying deals and invoices. A small sample is a reason to widen the confidence interval, not a reason to invent precision.

What should happen after an alert?

Open the source ledger, segment the variance, name the likely mechanism, run a contained test, and set a follow-up date. A red cell without an owner is decoration.

Are public benchmarks safe targets?

No. Benchmarks vary by business model, scale, billing terms, retention maturity, and service burden. Use them as context, then set internal alert bands from your cash window, contribution floor, capacity, and customer promise.

CONTROL ROOM ENTRY

Instrument the unit before scaling the story.

Make realized contribution, recovery speed, cohort quality, and cash timing visible every week.

Open The Control Sequence