# Meta Unit Economics > Calculate the unit economics beneath a Meta (Facebook and Instagram) ad account for ecommerce and content or creator clients, covering breakeven ROAS and CPA, target ROAS for a profit goal, MER, contribution margin, CAC, LTV, payback, and scaling headroom. This is the math layer that decides whether paid acquisition can work at all, separate from creative. Use whenever the question is whether Meta ads are profitable or can scale, what ROAS or CPA to aim for, why a campaign loses money despite good metrics, or whether an offer can support paid traffic. Trigger on breakeven ROAS, target ROAS, CPA target, MER, contribution margin, CAC, LTV, payback, AOV, "what ROAS do I need", "can I scale this", "am I actually profitable", or any Meta ad profitability, scaling-headroom, or offer-economics question. Ecommerce works in orders, AOV, and margin. Content clients work in cost per subscriber and value per subscriber. Do NOT use for creative testing (use meta-creative-engine) or Google Ads. - Skill: `kochellenk-afk/meta-unit-economics` (Agent Skill) - Install (CLI): `npx skillmds@latest add kochellenk-afk/meta-unit-economics` - Raw SKILL.md: https://api.skillmd.com/api/skills/kochellenk-afk/meta-unit-economics/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Marketing & Growth - Author: kochellenk-afk (https://skillmd.com/u/kochellenk-afk) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/kochellenk-afk/meta-unit-economics --- # Meta Unit Economics This is the math beneath the account. Creative decides which ad wins. Economics decides whether winning is even possible, what ROAS or CPA to aim for, and how far the account can scale before profit disappears. When meta-creative-engine's diagnostic finds that every creative metric is healthy but cost per result is still too high, the problem lives here, in the offer and the margin, not in the ad. Work in whichever unit matches the client. Ecommerce works in orders, AOV, and contribution margin. Content and creator clients work in cost per acquired subscriber or lead and the value that subscriber produces over time. The logic is the same in both: acquisition cost must sit below the value of what you acquire. ## Step 1: Confirm the client type and gather inputs **Ecommerce inputs:** AOV (average order value), COGS, shipping and fulfillment per order, payment processing fees, any other per-order variable cost, and if available: repeat rate, orders per customer per year, and the profit goal. **Content or creator inputs:** the acquisition target (subscriber, lead, install), and the value side: revenue per subscriber over time (ad or sponsorship revenue, product sales, email monetization), monetization rate, and time horizon. If a number is missing, ask for it or state the assumption you are using. Never invent margins silently. The output is only as honest as the cost inputs. ## Step 2: Pick the mode State which mode you are in. 1. **Breakeven and Targets.** Compute the ROAS and CPA the account must hit to break even, and the target to hit a stated profit goal. 2. **Scale Headroom.** Given current performance, calculate how far the account can scale before profit erodes, and where the ceiling is. 3. **Offer Economics.** When the numbers say paid acquisition cannot work at current AOV or margin, diagnose why and prescribe the levers that fix the economics. --- ## Mode 1: Breakeven and Targets **Ecommerce.** ``` Contribution margin per order (CM$) = AOV - COGS - shipping/fulfillment - fees - other variable Contribution margin % (CM%) = CM$ / AOV Breakeven CPA = CM$ (spend this much to acquire and you make zero) Breakeven ROAS = AOV / CM$ = 1 / CM% Target CPA (for net margin m as % of revenue) = CM$ - (m x AOV) Target ROAS (for net margin m) = 1 / (CM% - m) ``` Worked example: AOV 60, variable costs 30, so CM$ = 30 and CM% = 50 percent. Breakeven ROAS = 2.0, breakeven CPA = 30. To net 15 percent on revenue, target ROAS = 1 / (0.50 - 0.15) = 2.86, target CPA = 30 - (0.15 x 60) = 21. **Content or creator.** ``` Value per acquisition (VPA) = lifetime revenue per subscriber x contribution margin % Breakeven cost per acquisition = VPA Target cost per acquisition (for margin m) = VPA x (1 - m) ``` If cost per subscriber sits below VPA, acquisition is profitable. If not, it is buying attention at a loss and needs Mode 3. **Output template:** ``` ## Meta Economics: Breakeven and Targets [Client] Inputs: [AOV / variable costs / or VPA inputs, with any assumptions flagged] Contribution margin: [CM$ and CM%] Breakeven: ROAS [x] | CPA [amount] Target (for [m] profit): ROAS [x] | CPA [amount] Read: [one line on whether current performance clears the bar] ``` --- ## Mode 2: Scale Headroom Efficiency almost always drops as spend rises: the cheapest buyers are bought first, CPMs climb, and CPA rises with scale. The question is how much CPA can rise before profit hits zero. ``` Current CPA vs Breakeven CPA (= CM$) Headroom % = (Breakeven CPA - Current CPA) / Current CPA MER (blended) = total revenue / total ad spend (captures organic and halo) Breakeven MER = 1 / CM% (usually set the blended target below platform ROAS, because MER counts revenue that platform ROAS misses) ``` Interpretation: large headroom means push spend, efficiency can soften and still profit. Thin or negative headroom means the account is already near its ceiling. Scaling from there requires better economics (Mode 3) or better creative (hand back to meta-creative-engine), not more budget. If the client uses LTV, the affordable CPA rises to the lifetime contribution margin, not just the first-order margin. State the payback period so the client knows how long their cash is tied up: ``` LTV contribution = CM$ per order x orders per customer (over horizon) Affordable CAC (LTV basis) = LTV contribution Payback period = CAC / (CM$ per order per period) ``` **Output template:** ``` ## Meta Economics: Scale Headroom [Client] Current CPA [amount] vs Breakeven CPA [amount] | Headroom [%] Blended MER [x] vs Breakeven MER [x] LTV view (if used): affordable CAC [amount] | payback [time] Verdict: [scale / hold / fix economics first] ``` --- ## Mode 3: Offer Economics Use this when Mode 1 or 2 shows the offer cannot support paid traffic: breakeven ROAS is unrealistically high, CM% is thin, or CPA has no headroom. The fix is not a better ad. It is a better offer. Levers, in rough order of impact: - **Raise AOV.** Bundles, order bumps, post-purchase upsells, volume pricing, free-shipping thresholds. A higher AOV at the same margin percent lowers the required ROAS directly. - **Improve margin.** Reduce COGS, renegotiate fulfillment, cut discounting, fix payment fees. Every point of CM% lowers breakeven ROAS. - **Shift to an LTV model.** If first order cannot be profitable, build repeat purchase, subscription, or a back-end offer so acquisition is paid back over the customer lifetime rather than the first order. - **Reprice.** If the market allows, a higher price with the same variable cost widens CM$ fastest of all. For content and creator clients, the equivalent is raising value per subscriber: better monetization per subscriber, faster payback, or a higher-value acquisition target (a lead worth more than a follower). **Output template:** ``` ## Meta Economics: Offer Diagnosis [Client] Problem: [why the current offer cannot support paid, with the number] Required to work: [the CM% or AOV that would make target ROAS reachable] Levers, ranked: [the specific moves, each with its effect on breakeven ROAS] ``` --- ## Hand-offs - Creative metrics look weak (hook, hold, CTR): the problem is creative, not economics. Go to meta-creative-engine. - Landing page converts poorly: landing-page-match-scorer. - Positioning or offer narrative needs rebuilding: brand-narrative-positioning. ## Output rules - Never use em dashes. Use commas, periods, hyphens, or pipes. - Show the formula and the plugged-in numbers, not just the result, so the client can trust and re-run it. - Flag every assumption when an input is missing. Do not present an assumed margin as a known one. - Lead with the verdict (profitable, scalable, or not), then the math that proves it.