The Promotion Contribution Review Framework

PROMOTION CONTRIBUTION REVIEW — original Growthcraft synthesis
Decision / offer version / review date:
Owner of offer / finance / implementation / analysis / approval:
Currency / tax treatment / eligible visits / period:
Reference basket, price and product mix:
Qualifying order definition (at most one per eligible visit):

1. RESOLVE THE REAL OFFER
Product, order and shipping adjustments:
Eligibility, exclusions and combination settings:
Test-cart evidence (included, excluded, existing coupon, delivery):
Effective discounted basket revenue:
Unresolved checkout behaviour and owner:

2. RECONCILE COSTS
Shared variable cost per order:
Revenue-based fee and actual fee base:
Additional promotion cost per order:
Additional fixed campaign cost:
Return/refund maturity convention:
Costs deliberately excluded and why:
Evidence of no duplicated shipping, fees or media:

3. CALCULATE THE HURDLE
Reference and proposed conversion assumptions:
Reference / promotion contribution per order:
Contribution change at proposed conversion:
Required conversion / percentage-point change / relative change:
Status: representable / over 100% / nonpositive unit contribution / unsupported scope
Evidence that could make this hurdle realistic (unknown if absent):

4. ASSIGN THE NEXT DECISION
Hold / redesign offer / design bounded test / human review:
Required evidence, owner and due date:
Customer, margin, return and fulfilment guardrails:
Pre-agreed test design, budget cap and rollback conditions:
Unknowns and competing explanations:
Never translate a scenario into proof of lift or an automatic launch approval.

Use it before the promotional calendar becomes a commitment

A discount can increase orders while reducing contribution. This original operating framework helps a D2C growth lead, merchant and finance partner review that trade-off before implementation. It is a decision worksheet, not an empirically validated scoring system. Use it for a comparable basket and a clearly defined eligible visit population. The immediate job is to decide what evidence or redesign is needed—not to choose the largest discount that produces an attractive sales forecast.

Do not use the simple hurdle when product mix changes substantially, multiple orders per visit are material, traffic volumes differ, baseline unit contribution is nonpositive, or the objective is inventory liquidation regardless of current contribution. Those situations need an explicitly different model. A gift, bundle or free-delivery offer can be reviewed, but first translate the actual economic change into revenue and nonduplicated costs.

Prepare the evidence and responsibilities

The merchandising owner supplies the reference basket and exclusion rules. Implementation verifies checkout behaviour using authorised test-mode or non-purchasing test carts. Finance supplies the cost scope, fee base and return treatment. Analytics defines the eligible visit denominator and qualification event. The approver owns any eventual test budget. One person may hold several roles; no role should be silently replaced by the calculator.

Required inputs are aggregate assumptions, not customer records: basket revenue before and after the offer, variable costs, fee convention, additional campaign costs, comparable traffic, reference conversion and a proposed conversion scenario. If the only evidence is a coupon redemption count, record that limitation. Redeemers selected themselves into an offer and are not automatically a counterfactual group.

Four linked checks with explicit stop conditions

1. Offer contract: write down what the shopper actually pays. Verify included and excluded products, minimum spend, existing coupons and delivery changes. A failed or unverified cart combination is a hold, even if the spreadsheet looks profitable.
2. Cost contract: separate revenue-based charges, shared variable cost, extra cost per order and additional fixed campaign cost. Reconcile shipping subsidy and discounts so each appears once. An unknown cost is an unresolved input, not zero.
3. Economic hurdle: compare contribution at equal traffic. If promotion unit contribution is nonpositive, more promotional orders cannot rescue it under this model. If required conversion exceeds 100%, redesign. A feasible percentage merely removes an arithmetic impossibility; it does not establish demand.
4. Evidence handoff: identify what would justify a bounded test, with a stated owner and customer guardrails. A controlled experiment requires its own assignment, sample-size and analysis plan. Low-volume qualitative offer research answers a different question. Do not combine the two into a made-up confidence score.

Worked review: more orders, less contribution

Synthetic EUR scenario. A reference basket is 80, shared variable cost 38 and the simplified revenue fee 2.5%. Reference contribution is 40 per order. A 20% effective discount reduces basket revenue to 64; an additional delivery subsidy of 2 leaves promotion contribution of 22.40. At 10,000 visits and reference conversion of 2%, reference contribution is 8,000.

The proposed 3% conversion yields 300 expected orders, compared with 200 before. After 500 of extra campaign cost, promotion contribution is 6,220: a decline of 1,780 despite 50% more orders. Matching 8,000 requires approximately 3.7946% conversion, a 1.7946 percentage-point or 89.73% relative increase. This is not a required effect-size estimate for an A/B-test power calculation; it is the economic hurdle under the chosen assumptions.

The team has no causal response evidence and has not checked coupon stacking. Outcome: HOLD for checkout evidence and offer-response design, not launch. Finance can explore a smaller discount or a different subsidy, but every revised scenario must preserve its own assumptions. Do not pick the variant with the largest spreadsheet gain without asking whether the basket and traffic stay comparable.

Failure modes and limits

A blended AOV can hide changes in basket mix; an average percentage fee can hide fixed charges or taxes in the fee base. Returns can mature later than the campaign. Customers may bring purchases forward rather than add demand. Common acquisition spend is excluded because it cancels only when genuinely equal across scenarios. The framework does not measure brand effects, repeat purchase or inventory opportunity cost. Store the completed worksheet with the test plan and revisit it when the underlying assumptions change.

Source boundaries

Checked 25 September 2026. Shopify's current combination documentation distinguishes product, order and shipping discounts and their eligibility. Google's discount-event guide documents monetary item discounts and the paid-price field. These are implementation references, not validation of this original decision model or proof that a discount causes growth.