Commerce profit readiness audit

Profit Readiness Audit

Operating question: Do we have enough cost evidence to answer profit questions safely?

Run this as an evidence-first operating play. Gather the evidence below, separate exact / estimated / partial / unavailable data, apply the decision rules and vetoes, and never invent missing numbers. Be adversarial toward platform self-reported figures — trust commerce truth.

Evidence to gather

• Product catalogue with COGS — every active SKU, its unit cost, and its trailing revenue share. You need both % of SKUs with COGS populated and % of revenue covered by costed SKUs.
• Order / revenue data — orders by SKU for the window, so coverage can be revenue-weighted, not just counted.
• Payment-processing fees — actual fees charged by each processor over the window (or the contracted rate card if per-transaction data isn't exposed).
• Ad spend completeness — total spend per channel that ran, so you can confirm no spending channel is silently missing from the cost stack.

Optional, if available:

• Shipping / fulfilment cost feed — per-order shipping cost or 3PL/pick-pack invoice; usually the largest missing input.
• Returns / refund data — refund rate, refunded value, and whether returned goods re-enter sellable stock (recovered COGS) or are written off.
• Duties, FX, and inbound freight for cross-border or imported goods — quietly large on landed cost.
• Subscription / transaction app fees, chargebacks, and gift-card breakage — small individually, but they decide whether "net" is real net.

How to decide — in order

1. Measure COGS coverage by revenue, not by SKU count. "COGS on 90% of SKUs" can still be unsafe if the missing 10% are your bestsellers. Compute % of revenue covered. ≥95% revenue-covered → gross margin SAFE; 80–95% → PARTIAL (answer as a range); <80% → UNSAFE.
2. Name the gap, don't average over it. List the specific SKUs / categories with no cost. "Missing COGS = new-arrivals category, 22% of revenue" is actionable; "78% coverage" is not.
3. Walk the cost stack in profit order: Revenue → COGS → shipping/fulfilment → payment fees → ad spend → returns cost = contribution profit. A profit question is only as safe as its weakest required input.
4. Classify each profit question by the inputs it needs. Gross margin needs only COGS. Contribution profit needs COGS + shipping + fees + returns. Channel net needs all of that plus complete, correctly-attributed ad spend. Mark each SAFE / PARTIAL / UNSAFE by its weakest input.
5. For PARTIAL questions, derive the honest range, don't pick a point. Bound it: best case assumes missing-cost SKUs carry your highest known margin, worst case your lowest. If the range still keeps the decision on the same side of zero, it's decision-grade; if it straddles zero, it's UNSAFE for that decision.
6. Apply the vetoes, then publish the scorecard with the named owner and date for each blocking gap.

The prompt to run

You are my ecommerce finance analyst running the "Profit Readiness Audit" play.

PRE-FLIGHT: First list which required inputs I provided vs. missing. If a representative
sample of catalogue/order/cost data (COGS coverage and the cost-feed status) is missing,
STOP and return only (a) what's missing and (b) how to get it — never estimate it or proceed.

GOAL: do NOT compute my profit. Instead, tell me which profit questions I am currently
allowed to answer safely, given the cost evidence I have — and what is blocking the rest.

I will paste: COGS coverage (% of SKUs AND % of revenue, with the uncosted SKUs/categories
named), and the status of each cost feed — shipping/fulfilment, payment fees, ad spend per
channel, returns cost — marked Present / Partial / Missing with its source.

RULES:
- Weight COGS coverage by REVENUE, not SKU count. Missing cost on a bestseller is worse
  than missing it on 20 dead SKUs. Name the specific blocking SKUs/categories.
- Walk the stack in order: Revenue -> COGS -> shipping -> payment fees -> ad spend ->
  returns = contribution profit. A question is only as safe as its weakest required input.
- Classify each profit question SAFE (answer it), PARTIAL (answer only as a labelled range),
  or UNSAFE (refuse). Gross margin needs COGS; contribution needs COGS+shipping+fees+returns;
  channel net also needs complete ad spend.
- For PARTIAL, give the honest high/low range (missing-cost SKUs at my best vs worst known
  margin). Never collapse a range to a single number.
- Never state a profit number on partial cost data without labelling it partial and giving
  the range. Never fabricate a missing COGS, fee, or shipping cost.
- Every row carries a number, source, time window, and confidence level.

RETURN:
1. A 3-sentence executive read: what I can and cannot safely claim right now.
2. A readiness scorecard using EXACTLY this table header:

| Profit question | Inputs required | Inputs present | Verdict | Range if partial | Blocking gap |
|---|---|---|---|---|---|

   Use "—" for any cell you cannot fill from the evidence. Do not add or drop columns, and
   do not replace the table with prose.
3. The named SKUs / categories / feeds blocking the UNSAFE and PARTIAL answers.
4. The single highest-leverage gap to close first, with owner and ETA.

Decision rules

• SAFE (KEEP) — every required input is Present and current; for gross margin specifically, ≥95% of revenue is COGS-covered. The question can be answered as a point number.
• PARTIAL (WATCH) — the question's inputs are 80–95% covered by revenue, or one secondary input (returns, fees) is estimated. Answer only as a labelled range, never a point.
• UNSAFE (FIX) — a required input is Missing, or revenue coverage is <80%, or the honest range straddles the decision threshold. Refuse the answer and name what's blocked.
• REFRESH — a feed exists but has gone stale or drifted (e.g. COGS not updated since the last cost increase, payment rate card out of date). The input is recoverable without new infrastructure.
• KILL — retire a profit metric you keep reporting that can never be sourced safely (e.g. "true net by SKU" when no per-order shipping will ever exist) rather than reporting it wrong each month.
• Every recommendation must include a number, source, time window, and confidence level.

Vetoes — stop if any apply

• Never state a profit number on partial cost data without labelling it partial and giving the range. A point estimate over incomplete COGS is the single most damaging output of this play.
• Do not fabricate, interpolate, or category-average a missing COGS, fee, or shipping cost to "complete" the picture — a guessed cost is worse than a flagged gap.
• Do not report COGS coverage as a SKU-count percentage without the revenue-weighted percentage beside it.
• Do not let a downstream profit play (channel P&L, SKU cut, pricing) run on an input this audit marked UNSAFE.
• Do not collapse a PARTIAL range to its midpoint to make a slide look cleaner.
• Do not recommend writes, COGS backfills, feed changes, or catalogue edits without explicit human approval.

Output

A readiness scorecard: each profit question, the inputs it requires, what's actually present, the SAFE/PARTIAL/UNSAFE verdict, and the named blockers. Plus a 3-sentence executive read.

Minimum table columns:

Profit question	Inputs required	Inputs present	Verdict	Range if partial	Blocking gap
Gross margin (blended)	COGS coverage	96% rev-covered	SAFE	—	None

---

Run this on your live store data in one prompt — start a free 14-day trial of ShopMCP (no credit card). It connects your store, ads, analytics, email/SMS, and finance and runs this play read-only by default, applying changes only on your approval.