Most marketplace dashboards are very good at telling you what happened. Amazon.de revenue was up 18%. bol.com conversion dropped. Walmart ad spend paced ahead of plan. Shopify had a strong email weekend. Lovely. Also not enough.
The question for a brand owner is not “which channel grew?” The better question is: which channel created contribution margin we can safely reinvest? That is a different operating conversation, especially once you sell the same product family across Amazon, bol.com, Mirakl retailers, Shopify, Walmart, TikTok Shop or Google Shopping.
The named mistake I see is dashboard applause: the team celebrates the biggest revenue bar, then gives that channel more stock, more budget and more attention before checking the cost layers underneath. A channel can win the revenue chart and lose the profit meeting. It happens more often than people admit.
My stance: multi-channel analytics should not start with a channel ranking. It should start with a contribution margin waterfall. A waterfall shows how gross revenue turns into net contribution after discounts, marketplace commission, fulfilment, payment fees, returns, ad spend, storage, content or creator costs, and operational exceptions. It turns a dashboard from a scoreboard into a decision tool.
This guide is for marketplace brand owners doing roughly €1.5K+ monthly retail media spend or 1K+ orders a month. At that stage, spreadsheets still feel manageable, but the decision risk is already real: one wrong channel gets budget, one profitable SKU is starved of stock, one “great” promotion quietly funds the marketplace instead of the business.
What the current analytics advice gets right
The existing software content is useful. Jungle Scout and Helium 10 explain why Amazon sellers need sales analytics, profit tracking, COGS, fees, PPC spend, refunds and inventory visibility instead of manually rebuilding numbers in spreadsheets. sellerboard focuses heavily on accurate Amazon profit analytics, including FBA and FBM shipping costs, refunds, Amazon fees, COGS, indirect expenses and PPC. MerchantSpring’s Amazon marketplace analytics guide correctly pushes sellers beyond sales tracking into conversion, pricing, ad effectiveness, inventory and profit after fees, refunds, ads and COGS. DataHawk’s positioning is also moving in the right direction: unified analytics, SKU-level profitability, ad metrics, inventory signals and AI diagnosis across Amazon, Walmart and other marketplaces.
Reddit threads show the operator pain behind those landing pages. Sellers ask how to calculate profit per SKU, how to treat inbound shipping in COGS, how to reconcile Amazon fees, why PPC is eating margin, and whether any report gives the actual per-ASIN profit after fees and ads. The recurring theme is not a lack of metrics. It is a lack of trust in the final number.
The gap is that most advice stops at “track true profit” or “use a profit dashboard”. That is important, but it misses the multi-channel decision layer. Brand owners do not only need to know whether Amazon is profitable. They need to decide whether Amazon.de, bol.com NL, Walmart US, Shopify and a Mirakl retailer deserve the next €5,000 of stock, the next €1,500 of ad spend, the next price test or the next product content sprint.
That is why the waterfall matters. It does not just calculate margin. It explains where margin disappears by channel.
What a contribution margin waterfall actually shows
A contribution margin waterfall starts with gross revenue and walks down to the money left to cover fixed costs and profit. For marketplace analytics, the clean version has nine layers:
- Gross revenue: item price before discounts, returns and channel adjustments.
- Promotions and discounts: coupons, deals, vouchers, price reductions and marketplace-funded versus seller-funded discounts.
- Net sales: revenue after commercial reductions and cancellations.
- Marketplace and payment fees: referral commission, platform fees, payment processing and category-specific charges.
- Fulfilment and shipping: FBA, LVB, WFS, 3PL picking, outbound shipping, packaging and inbound allocation.
- COGS: product cost, landed freight, duties and any unit-level packaging cost you control.
- Returns and refunds: refund value, return handling, unsellable stock, return shipping and expected loss rate.
- Retail media and acquisition spend: Sponsored Products, Sponsored Brands, bol Ads, Walmart Connect, Google Shopping, Meta or creator commission where it drives marketplace demand.
- Contribution margin: the remaining euro amount and percentage, ideally at SKU, parent product and channel level.
The trick is not the formula. The trick is consistency. If Amazon includes a fee timing lag, bol.com reports fulfilment differently, Shopify sees payment fees immediately and Walmart settles returns later, your waterfall must normalize those differences before you compare channels. Otherwise you are not doing analytics. You are comparing accounting dialects.
Example 1: the Amazon hero SKU that looked better than it was
Imagine a Dutch cookware brand selling the same induction pan set on Amazon.de, bol.com NL and Shopify. In July, the pan set looks like an obvious Amazon winner:
- Amazon.de: €62,400 gross revenue from 1,200 orders
- bol.com NL: €38,700 gross revenue from 860 orders
- Shopify: €21,600 gross revenue from 360 orders
If the team ranks channels by gross revenue, Amazon gets the next purchase order and the next ad budget increase. The waterfall tells a calmer story.
Amazon.de has a €52 selling price. After an average €3 coupon, €7.80 referral fee, €6.40 FBA fulfilment, €18.50 landed COGS, €2.60 expected return cost and €8.20 ad spend per order, contribution margin is €5.50 per order. On 1,200 orders, that is €6,600 contribution.
bol.com sells at €45. After €5.85 commission, €5.10 LVB fulfilment, €18.50 COGS, €1.75 return cost and €3.90 Sponsored Products spend per order, contribution margin is €9.90 per order. On 860 orders, that is €8,514 contribution.
Shopify sells at €60. After €2.10 payment fees, €6.80 shipping, €18.50 COGS, €2.40 return cost and €14.00 blended Meta and email acquisition cost per order, contribution margin is €16.20 per order. On 360 orders, that is €5,832 contribution.
The revenue dashboard says Amazon is 61% bigger than bol.com. The contribution waterfall says bol.com created €1,914 more contribution. That changes the decision. Amazon is not bad, but it has a coupon-and-ad-spend problem. bol.com deserves stock protection. Shopify deserves a retention test before paid scaling. The next action is not “scale Amazon”. It is “fix Amazon’s commercial layers before giving it more oxygen”.
This is exactly where FiveX marketplace analytics should sit in the workflow: connect sales, ad spend, fees, fulfilment, COGS and returns into one SKU-level view so the operator can see which layer changed the answer.
Example 2: the French marketplace that needed a price fix, not an ad cut
Now take a sports nutrition brand selling a protein shaker across Amazon.fr, a Mirakl-powered French retailer and its own Shopify store. The performance meeting starts with a familiar complaint: “France is inefficient. ROAS is only 2.4. Cut spend.”
The SKU sells 2,000 units a month in France at an average gross price of €18. The blended ad spend is €7,500. At first glance, that feels heavy. But the waterfall separates the problem:
- Gross revenue: €36,000
- Seller-funded discounts: -€3,600
- Marketplace and payment fees: -€4,950
- Fulfilment and packaging: -€6,200
- COGS: -€12,000
- Returns and damage allowance: -€900
- Retail media spend: -€7,500
- Contribution margin: €850
Only €850 contribution on €36,000 revenue is not healthy. But the named mistake would be cutting ads first because ROAS looks ugly. When the team compares the same SKU across channels, they find the French marketplace is discounting €1.80 per unit to match a competitor while Shopify holds price and still converts through bundles. The ad spend is not the only leak. The price architecture is.
A cleaner test is: remove the €1.80 discount for two weeks, keep Sponsored Products spend flat, and accept a 12% unit drop if contribution per unit rises. If units fall from 2,000 to 1,760 but discount leakage improves by €3,168, contribution can move from €850 to roughly €3,490 even before campaign optimization. That is a price-and-positioning decision, not a panic ad cut.
FiveX helps here by combining repricing guardrails with analytics. The question is not just “can we win the offer?” It is “can we win the offer while preserving contribution margin after fees, fulfilment, returns and ads?” Different question. Much better answer.
Example 3: the US launch that needed an inventory gate
A home storage brand launches on Walmart US while Amazon.com remains the main volume channel. In week three, Walmart’s dashboard looks exciting: 420 orders, €24,780 equivalent gross revenue and 18% week-on-week growth. The team wants to move €2,000 more ad budget into Walmart Connect.
The waterfall says “not yet”. Contribution margin is positive at €4.80 per order, but stock cover is only 11 days because the next inbound shipment is delayed. If ads scale by 40%, the SKU will stock out before replenishment lands. On Walmart, that means lost momentum. On Amazon, reallocating stock away from FBA can also weaken the main channel.
The decision becomes an inventory gate: keep Walmart spend flat until stock cover is above 21 days, protect Amazon’s baseline availability, and use the waiting period to improve content and reviews. The waterfall is still useful, but only when it is paired with stock. Margin you cannot fulfil is theoretical margin. Very beautiful in a spreadsheet. Very useless in operations.
This is the third FiveX hook: connect contribution margin to stock management and data exports. Finance sees the margin, operators see the stock constraint, and the ad owner sees why scaling is paused without needing a dramatic Slack debate.
The five decisions your waterfall should trigger
A good contribution margin waterfall is not a monthly finance artifact. It should trigger weekly operating decisions.
1. Scale
Scale when contribution per order is positive, stock cover is safe, return rate is stable and paid spend is not masking an organic decline. The rule can be simple: increase budget only when the SKU has at least 20% contribution margin, 21+ days of stock cover and no return-rate spike above the last eight-week average.
2. Fix price
If fee and fulfilment layers are stable but contribution drops because discounts rise, fix price architecture before campaign structure. Many teams waste hours on bid changes when the real issue is a €2 voucher that nobody owns.
3. Fix operations
If return cost or fulfilment cost is the leak, ads are not the first lever. Improve product content, packaging, delivery promise, variant mapping or marketplace fulfilment choice. A 4-point return-rate reduction can beat a heroic ROAS improvement.
4. Reallocate stock
If two channels have similar revenue but one creates twice the contribution per unit, stock should not be allocated by habit. The waterfall gives supply chain a commercial reason to prioritize one channel over another.
5. Quarantine
Some SKUs should be held out of scale. Not deleted, not ignored, just quarantined until the leak is named. A product with €40,000 monthly revenue and -€2,000 contribution is not a growth engine. It is a very busy hole.
How to build the dashboard without creating spreadsheet theatre
Start with one parent SKU family across three channels. Do not try to rebuild your entire company P&L in week one. Pick a product where sales, ads and stock decisions happen every week. Then build the waterfall with the same definitions per channel.
Use order-level or SKU-day granularity where possible. Map channel SKUs to one internal product ID. Store COGS and landed cost with effective dates, because product cost changes over time. Separate seller-funded discounts from marketplace-funded promotions. Allocate ad spend by SKU where the platform allows it, and use a documented rule where it does not. Treat returns as expected cost if the final settlement lags. Then compare both euro contribution and contribution percentage.
The operator view should fit on one screen: revenue, units, contribution, contribution margin, ad spend, return cost, stock cover and the biggest waterfall leak versus the previous period. If the dashboard needs twenty tabs to explain itself, it will become another reporting museum.
In FiveX, this is the point of bringing marketplace integrations, advertising data, product profitability and operational signals into one cockpit. The software should not merely report that margin dropped. It should help the team see whether margin dropped because ad spend increased, stock forced a fulfilment change, returns rose, price was discounted, or a fee category changed.
The practical weekly cadence
Run the waterfall review once a week, preferably before budget and replenishment decisions. Keep it boring. Boring is underrated in profitable ecommerce.
- Monday: review the top 20 SKU-channel combinations by revenue and the top 20 by negative contribution movement.
- Tuesday: decide scale, fix, hold or quarantine for each exception.
- Wednesday: change bids, budgets, prices, content or fulfilment rules.
- Friday: check whether the leak moved. Do not wait for month-end finance to discover what operators could have fixed three weeks earlier.
The trade-off is speed versus accounting perfection. Your waterfall will never be as final as the closed monthly books. That is fine. Operators need a decision-grade number, not a museum-grade number. If the number is directionally reliable, consistently defined and close enough to prevent bad scaling decisions, it has already paid for itself.
Bottom line
Multi-channel analytics becomes powerful when it stops asking “which channel sold most?” and starts asking “which channel created contribution we can safely grow?” A contribution margin waterfall gives brand owners that answer. It exposes the difference between revenue, profitable revenue and scalable profitable revenue.
The best marketplace teams do not use waterfalls to admire margin erosion. They use them to make sharper decisions: where to scale spend, where to fix price, where to protect stock, where to repair operations and where to pause until the economics make sense.
If you want that view without rebuilding spreadsheets every week, FiveX connects marketplace analytics, advertising, repricing, product profitability, stock and exports in one operating cockpit. Not to create prettier charts. To stop dashboard applause from turning into expensive growth.