Deterministic Cart Simulation for Better Promotion QA

# Deterministic Cart Simulation for Better Promotion QA Promotion QA often begins with an example cart assembled by hand. That is useful, but it can be hard to reproduce: products change, sessions persist, customer histories differ, and third-party extensions add conditions. A deterministic cart probe gives the team a stable diagnostic step before it spends time on full checkout testing. The idea is simple. Describe a hypothetical cart with a small, bounded set of inputs, then evaluate the published coupon rules against that description without creating a WooCommerce cart. ## Choose inputs that explain eligibility A practical probe can use: - cart subtotal; - eligible item quantity; - product IDs; - category IDs; - an optional customer email pattern; and - whether a sale item is present. Those fields cover common native WooCommerce coupon boundaries. The evaluator should also consider expiration, total usage exhaustion, minimum and maximum spend, included and excluded products/categories, and sale-item rules. Every ineligible result should say why. The same coupon snapshot, probe, thresholds, and evaluation time must produce the same output. That property is valuable during campaign review because two people can discuss the same result instead of comparing different carts. ## Estimate exposure conservatively For triage, a percent coupon can be estimated against the probe subtotal, a fixed-cart coupon at its configured amount, and a fixed-product coupon against the bounded quantity. Cap combined exposure at subtotal. Separate coupons that permit stacking from individual-use coupons so the most concerning concurrent set is visible. This estimate is not checkout parity. It cannot know exact line allocation from a list of IDs, and it should not pretend to model taxes, shipping, currency conversion, per-customer redemption history, or third-party rule filters. Honest boundaries make the result more useful, not less. ## Turn probes into a test ladder Use deterministic results to decide what to test next: 1. Run probes for representative low, median, and high cart values. 2. Add products/categories from campaigns that overlap in time. 3. Sort for the highest combined stackable estimate. 4. Build real carts for those combinations in a disposable or approved test store. 5. Verify the authoritative WooCommerce total and record any extension-specific behavior. [Vamao WooCommerce Coupon Exposure OS](/products/vamao-woocommerce-coupon-exposure-os) provides the diagnostic layer: a read-only snapshot, explainable findings, bounded simulation, and CSV export. It deliberately leaves final validation to WooCommerce checkout. That separation prevents a common tooling mistake. A simulator should help operators find questions; it should not silently become another promotion engine with its own version of the truth.