--- name: build-an-ltv-cac-model title: Build an LTV to CAC Model description: An LTV to CAC prompt and ecommerce unit economics model that shows how to calculate LTV to CAC and set the ceiling on what you can afford to pay for a customer. cluster: ecommerce version: 1.1.0 --- # Build an LTV to CAC Model This prompt turns a pile of order data and cost lines into a defensible customer economics model. It computes lifetime value on a contribution margin basis, picks the correct cost per acquisition definition for the decision you are making, sets a profit maximizing target ratio, and stress tests the whole thing against your cash position so you never scale into a liquidity trap. The output is a model you can hand to a finance partner and a single acquisition ceiling your marketing team can act on daily. ## When to use this → You want to know how much you can afford to pay to acquire a customer, and every current estimate is a guess → Acquisition cost is rising and you need to prove whether higher spend still pays back → You are shifting from judging each order in isolation to judging profit per customer over the full relationship → You are raising capital or planning a budget and need per unit fundamentals that hold up to scrutiny ## The prompt ```text You are a senior ecommerce unit economics analyst. Build a rigorous LTV to CAC model, choose the right cost definitions, recommend a profit maximizing target, and flag cash flow risk. Be precise and conservative; never present a number you cannot defend. Show your work: for every derived number, write the formula with my inputs substituted so I can recompute it line by line. Label any value you estimated rather than received as ESTIMATE. CONTEXT: → Business model: {{BUSINESS_MODEL}} (subscription, one time purchase, or mixed) → First order AOV, net revenue: {{FIRST_ORDER_AOV}} → Per order costs: {{COGS}}, {{LOGISTICS_FULFILLMENT}}, {{PAYMENT_FEES}} → Repeat behavior: {{REPURCHASE_OR_RETENTION}} (repurchase or churn rate, cohort curve if available) → Marketing: {{MEDIA_SPEND}}, {{CREATIVE_AND_AGENCY}}, {{FIXED_TEAM_COSTS}} → New customers in the period: {{NEW_CUSTOMERS}} → Cash constraints: {{CASH_IN_BANK}}, {{SUPPLIER_PAYMENT_TERMS}}, {{INVENTORY_LEAD_TIME}} → Horizon and required return: {{HORIZON}}, {{REQUIRED_RETURN}} STEP 0. CLARIFY FIRST. Before any math, audit the context block. If first order AOV, any of the three cost lines, repeat data, or the cash position is missing or ambiguous, list every missing input in a single message, state what each one is needed for, then stop and wait for my answers. Do not proceed on placeholders. Never fabricate a value curve; if repeat data is genuinely unavailable, model only first order economics and label the result as first order only. STEP 1. BUILD LTV ON CONTRIBUTION MARGIN, NEVER ON REVENUE. Walk the P and L: Gross revenue minus returns and taxes equals Net revenue. Net revenue minus COGS minus logistics and fulfillment minus payment fees equals Contribution Margin 2 (CM2). LTV is accumulated CM2 per customer across the relationship, not revenue. Reject any revenue based LTV. Keep marketing out of LTV; it belongs in CAC, since spend is front loaded on the first order while repeat purchases need little spend. STEP 2. CONFIRM A RELATIONSHIP MODEL APPLIES. Use LTV to CAC only when the product supports repeat purchase. For a genuine one and done product, stay on per order economics and say so. STEP 3. CHART THE VALUE CURVE, NOT JUST THE TOTAL. State total LTV, lifetime length, and the shape of value accrual (front loaded and concave, linear, or back loaded). Two customers with identical total LTV differ in cash impact: front loaded value and shorter lifetimes reach breakeven sooner. Derive the curve from cohort data, not intuition. STEP 4. CHOOSE THE CAC DEFINITION THAT MATCHES THE DECISION. Compute all three and label which one each downstream statement uses. → Media CAC (media spend only): optimize channels, campaigns, ads. → Variable CAC (media plus creative plus agency): the primary number, for budgeting and the headline ratio. → Fully loaded CAC (variable plus fixed team costs): check economies of scale; it should fall as a share of revenue as you grow. STEP 5. FIND THE LIQUIDITY GAP. Compare first order CM2 against variable CAC. If CAC exceeds first order value, state how many orders and days to break even. On a bootstrapped business, liquidity matters as much as total profitability; the best operators manage to a target days to break even per customer, not a profit target. STEP 6. SET THE TARGET RATIO FOR MAXIMUM TOTAL PROFIT, NOT THE HIGHEST RATIO. Optimize total absolute bottom line profit over the horizon (unit economics times volume acquirable at that CAC), subject to a liquidity cushion you choose deliberately. Model two scenarios (high ratio with low spend, lower ratio with higher spend) and compare total profit. A worse ratio that acquires far more customers can win. STEP 7. STRESS TEST TIMING AND CASH. Timing dominates the ratio: pushing value into the future can turn a profitable configuration unprofitable at an unchanged ratio. Test two levers, raising the target ratio (lower CAC) and improving supplier terms so you pay for inventory after selling it. Never run every buffer at zero. STEP 8. CONVERT THE MODEL INTO ONE OPERATIONAL NUMBER. Give the team a single maximum CAC to act on: "spend up to X per new customer." CAC is the lever they steer daily; LTV is too abstract. Always judge acquisition on CAC relative to LTV or time to breakeven, since similar CACs hide very different LTVs. OUTPUT FORMAT: 1. CM2 and LTV calculation, line by line, each line showing the formula with inputs substituted. 2. All three CAC values, primary one flagged. 3. Value curve (shape, length, breakeven in orders and days). 4. Scenario table: ratio, CAC, customers acquirable, total profit, liquidity gap. 5. Recommended target ratio and the single max CAC number. 6. Cash flow guardrails and top risks. 7. FACT CHECK LIST: every assumption and every value marked ESTIMATE, with the output number it moves and the direction it moves it, so I can verify each one before acting. For retention driven models, call out that small retention errors flip the outcome. SELF CHECK: → LTV is CM2 based, not revenue based. → Marketing excluded from LTV, captured in CAC. → You did not default to the highest ratio. → The plan is liquidity safe given cash and supplier terms. → Every derived number shows its formula; every estimate carries the ESTIMATE label and appears in the fact check list. ``` ## Prompt versions The standard prompt above works on any model. Use these variants when you want a different tradeoff. ### Frontier model version Built for the most capable models (Claude Opus and beyond). States the goal, constraints, and quality bar up front, then trusts the model to choose its path. ```text You are a senior ecommerce unit economics analyst. Build me a defensible LTV to CAC model and resolve it into one number my team can act on: the maximum I can afford to pay per new customer. Lead your output with that ceiling and a one line verdict on whether current spend pays back, then show the model beneath it. CONTEXT: → Business model: {{BUSINESS_MODEL}} → First order AOV, net revenue: {{FIRST_ORDER_AOV}} → Per order costs: {{COGS}}, {{LOGISTICS_FULFILLMENT}}, {{PAYMENT_FEES}} → Repeat behavior: {{REPURCHASE_OR_RETENTION}} → Marketing: {{MEDIA_SPEND}}, {{CREATIVE_AND_AGENCY}}, {{FIXED_TEAM_COSTS}} → New customers in the period: {{NEW_CUSTOMERS}} → Cash: {{CASH_IN_BANK}}, {{SUPPLIER_PAYMENT_TERMS}}, {{INVENTORY_LEAD_TIME}} → Horizon and required return: {{HORIZON}}, {{REQUIRED_RETURN}} Load bearing method (these are non negotiable): → Build LTV as accumulated Contribution Margin 2 (net revenue minus COGS, logistics, and payment fees), never on revenue. Marketing stays out of LTV and lives in CAC, because spend is front loaded on the first order. → Use LTV to CAC only if the product genuinely repeats; for a one and done product, stay on per order economics and say so. → Compute all three CAC definitions (media, variable, fully loaded), flag variable CAC as the headline ratio driver, and use fully loaded CAC to test whether unit economics improve with scale. → Read the value curve, not just the total: two customers with equal LTV differ in cash impact when one is front loaded and shorter lived. Derive shape from cohort data. → Find the liquidity gap (first order CM2 versus variable CAC) and express breakeven in orders and days. On a cash constrained business, days to breakeven can outrank total profit. → Choose the target ratio that maximizes total absolute profit over the horizon, not the highest ratio; a worse ratio that acquires far more customers can win. Never run buffers at zero, and account for supplier terms and inventory lead time. Quality bar: every derived number shows its formula with my inputs substituted so I can recompute it line by line; every value you estimated is labeled ESTIMATE and appears in a closing fact check list naming the output it moves and the direction; and the whole model is liquidity safe given my cash and supplier terms. Do not fabricate a value curve, an AOV, or a retention rate. If first order AOV, any cost line, repeat data, or the cash position is missing or ambiguous, ask one focused round of questions before computing anything rather than guessing. Do not pad with generic advice. For retention driven models, call out that small retention errors can flip the outcome. ``` ### Quick version Five lines or fewer, for when speed matters more than rigor. ```text Give me the maximum I can afford to pay to acquire a customer, stated as one CAC ceiling. Business: {{BUSINESS_MODEL}}. First order net AOV {{FIRST_ORDER_AOV}}; per order costs {{COGS}}, {{LOGISTICS_FULFILLMENT}}, {{PAYMENT_FEES}}; repeat {{REPURCHASE_OR_RETENTION}}. Build LTV on Contribution Margin 2 (never revenue) with marketing excluded, then set the ceiling. Show each formula with my numbers, and if repeat data is missing, ask before guessing. ``` ## How to customize → `{{BUSINESS_MODEL}}`: subscription, one time purchase, or mixed. For example "replenishment subscription, monthly billing." → `{{FIRST_ORDER_AOV}}`: average net revenue on the first order, after returns and taxes. For example "$62 net." → `{{COGS}}`, `{{LOGISTICS_FULFILLMENT}}`, `{{PAYMENT_FEES}}`: the three variable cost lines hit on every order. For example "COGS $22, fulfillment $8, processing 2.9% plus $0.30." → `{{REPURCHASE_OR_RETENTION}}`: churn rate for subscriptions or repurchase rate for one time models, plus a cohort curve if you have one. For example "8% monthly churn" or "50% repurchase by day 120." → `{{MEDIA_SPEND}}`, `{{CREATIVE_AND_AGENCY}}`, `{{FIXED_TEAM_COSTS}}`: the three cost layers that build the three CAC definitions. → `{{CASH_IN_BANK}}`, `{{SUPPLIER_PAYMENT_TERMS}}`, `{{INVENTORY_LEAD_TIME}}`: the liquidity inputs. For example "net 60 terms, 90 day inventory lead time." → `{{HORIZON}}`, `{{REQUIRED_RETURN}}`: the window you optimize over and the return multiple that makes tying up capital worthwhile. ## What good output looks like → LTV is computed from Contribution Margin 2 line by line, every line showing its formula with your inputs substituted so you can recompute it, with marketing excluded and any revenue based LTV explicitly rejected → All three CAC definitions appear, with variable CAC flagged as the ratio driver and fully loaded CAC used to test scale → The recommendation names a target ratio chosen to maximize total profit, backed by a scenario comparison, not the highest ratio available → A liquidity gap is quantified in orders and days, and the plan is checked against cash in bank and supplier terms → The model resolves into one number a marketer can act on: a maximum CAC per new customer → The output ends with a fact check list: every assumption and estimated input, the number it moves, and the direction, ready for you to verify before anyone spends against the model → If required inputs were missing, the model asked for all of them in one message before computing anything, instead of filling gaps with invented values ## Related prompts → [Design Offers That Raise Order Value](../ecommerce/design-offers-that-raise-aov.md): once you know your economics, raise the LTV side of the ratio through offer and basket design. → [Forecast Revenue and Inventory From Cohorts](../ecommerce/forecast-revenue-and-inventory.md): feed the same cohort curves into a forward forecast of revenue and stock needs. → [Optimize Product Discovery](../ecommerce/optimize-product-discovery.md): lower the CAC side of the ratio by improving how customers find and choose your products.