Subtract COGS and Your No. 1 Product Changes — But Not If You Stop at Gross Profit

Ranking products by revenue and by gross profit produced what was essentially the same list (rank correlation 0.99). The order only flipped once we went down to contribution margin (0.58). The No. 2 product by revenue came dead last out of 14 on contribution margin, and the No. 3 product by revenue ranked No. 1. What made the difference wasn’t COGS — it was the ad spend attached to each product.

Public research data + sample data from a commerce brand · 12-month window · approx. 270,000 orders / approx. 134,000 purchasers · 187 SKUs with sales. For every section, we also note the Laplace screen and settings where you can see that number.

The figures in this report are not real transaction data; they are illustrative values generated to reflect the distribution of Laplace benchmark metrics. They do not represent any specific customer, so please use them to understand trends rather than as absolute values.

---

PART 01 · First — Which screens to open

You don’t need to look at every menu. We picked only the screens that produce the numbers in this report.

Only one row needs to be built from scratch

---

PART 02 · Half of your revenue ranking barely makes any profit — the numbers research has confirmed

Before looking at our own data, let’s set a baseline. How unevenly profit is distributed across products has been measured at scale.

In other words, “the products that sell well” and “the products that make money” are two different lists. The problem is that most brands only have the first one.

So far, these are other people’s numbers

Next screen — Research only goes as far as “each SKU contributes differently to profit.” It doesn’t tell you how far down the revenue ranking you need to go before that difference shows up.

---

PART 03 · Priority 1 — Rankings flip at contribution margin, not at gross profit

Brand A is a Korean skincare brand. It sells serums, creams and ampoules through its own store, major platforms and marketplaces. Over 12 months, 187 of its 199 registered SKUs generated sales.

Before reading the numbers

Where to look in Laplace — P & L Analysis › Margin Rate / Contribution Margin Rate

How to set it up

1. Set the date range to 12 months — so that a single seasonal product can’t shake up the ranking.

2. Set the dimension to Product. If you choose Option, each quantity option becomes its own row and the ranking gets fragmented.

3. Turn on all three metrics: revenue · gross profit · contribution margin. With only one of them on, you won’t see this section’s conclusion.

4. Confirm that you’re using net sales after refunds and claims. Revenue before those adjustments overstates products with high return rates.

Revenue index: No. 1 product = 100. Contribution margin rate = 100 − COGS rate − fee rate − shipping rate − ad spend rate (based on net sales excluding VAT).

How to read the numbers

• The gross profit ranking is the same list as the revenue ranking. Rank correlation 0.99, average shift 0.3 places. That’s because COGS rates don’t vary enough to change the order — in this table they cluster between 22.6% and 39.1%. If you stop at gross profit and say “we look at things on a profit basis,” you are effectively still looking at revenue.

• The first flip happens at contribution margin. Rank correlation 0.58, average shift 3.1 places. The No. 2 product by revenue drops to No. 14 (last) on contribution margin, while the No. 14 product by revenue climbs to No. 8.

• What flipped the ranking wasn’t COGS — it was ad spend. The same table contains a product whose ad spend is 8.8% of net sales and another at 67.5%. The max–min gap in COGS rate is 16.5pp, while the gap in ad spend rate is 58.7pp.

• Loss-making products sit near the top of the revenue ranking. The No. 2 and No. 13 products by revenue have negative contribution margins. Looking only at the revenue ranking, both of them are “doing well.”

Next screen — We now know the ranking flips. What we can’t see yet is where each 100 won of revenue flows to make that happen.

---

PART 04 · Breaking down 100 of revenue — how gross profit and contribution margin differ

Where to look in Laplace — P & L Analysis › Margin Rate / Contribution Margin Rate

The difference between the two metrics — how much you subtract

How to read the numbers

• Overall, gross margin is 68.6% and the contribution margin rate is 21.1%. The 47.5pp in between goes to fees (12.3%), shipping (3.4%) and ad spend (31.8%). The single biggest cost isn’t COGS — it’s ad spend.

• The same product has the opposite cost structure on different channels. On the own store, fees are thin and ad spend is heavy; on major platforms, fees are heavy and ad spend is thin. The combined contribution margin rates may look similar, but the items you can cut are different.

• A product with a 0.6% contribution margin rate ranks No. 9 in revenue. Ad spend climbed to 67.5% and ate up almost all of its 77.4% gross margin. It succeeded at selling, but nothing was left over.

Next screen — Every number so far rests on the assumption that the table is accurate. In reality, it’s more common for the table itself to be skewed.

---

PART 05 · Pitfalls — skip these three checks and every table above is wrong

Please follow this order. If you put this section off, you’ll end up trusting the previous section’s numbers at face value.

Pitfall ① The same product is split across multiple product names

Different product names per channel, quantity options (1-pack/2-pack/4-pack) and pre/post-renewal labels are each captured as separate products. At Brand A, one product was split across 19 product names, and their contribution margin rates ranged from -31.4% to +72.5%. Combined, it’s 8.8% — and looking at that single line alone, nothing seems wrong.

How to check

1. In Data Management › Base product, see how many products the same item is registered as.

2. Look at the product axis (Base product) and the sales axis (channel products) separately. Roll up on the product axis and you see how good the product really is; spread out on the sales axis and you see problems with how it’s sold.

3. Look at quantity options separately in Product Analysis › By Option. 2-packs and 4-packs are usually repeat-purchase volume sold without ads, so their contribution margin rates are much higher.

Pitfall ② If you don’t attach ad spend to products, you get a fake contribution margin

If you only manage ad spend as a lump sum, the PART 03 table becomes identical to the gross profit table — you end up calling a table with a 0.99 rank correlation “contribution margin.” Conversely, if you look at contribution margin by channel without knowing the allocation rules, you’ll be wrong in the opposite direction.

The small marketplace’s −91.6% does not mean the channel is bad

Pitfall ③ Products without COGS attached have inflated profits

Products whose COGS isn’t matched are calculated as having zero COGS and show up with a 100% gross margin. If such a product is a top seller, the entire profit margin gets inflated.

It doesn’t have to be perfect

Next screen — The table is now trustworthy. Next, we decide which box each product goes into.

---

PART 06 · Decision ① Reshaping the portfolio — the 4-quadrant view

Where to look in Laplace — Custom Dashboard › Product 4-Quadrant Chart

Don’t cut everything just because its contribution margin rate is low

Next screen — We’ve decided which products deserve more ad spend. What we don’t know yet is how much is safe to put in.

---

PART 07 · Decision ② Ad budget — break-even ROAS differs by product

Where to look in Laplace — Ad Analysis › By Ad Channel · By Campaign

Break-even ROAS = 100 ÷ (100 − COGS rate − fee rate − shipping rate)

How to read the numbers

• A blanket target like “3x ROAS for every product” is wrong. In this sample, the baseline ranges from 1.47x to 2.82x. Give everything a single target and you’ll cut ads on some products that are actually profitable while continuing to burn money on others that are losing it.

• Channels shouldn’t share one target either. Own store 1.70x, major platform 2.29x — because their fee structures differ. Giving every channel the same ROAS target is the most common mistake.

• If you rank by contribution margin instead of ROAS, the order shifts a little. But when COGS rates are clustered in a narrow band, as at this brand, the two orders don’t differ much. The wider the spread of COGS rates at a brand, the more ROAS and actual profit diverge in ranking.

Next screen — The ad budget is set. The remaining lever is discounting.

---

PART 08 · Decision ③ Discount limits — the ceiling is the contribution margin rate

A blanket discount like “20% off everything” pushes thin-margin products into the red. You need contribution margin rates by product to set discount limits by product.

Discounting is a far more expensive tool than you think

---

PART 09 · Decision ④ SKU rationalization — raise margins while protecting revenue

Where to look in Laplace — P & L Analysis › Margin Rate / Contribution Margin Rate

How to read the numbers

• 21 SKUs generate 80% of contribution margin — that’s 11.2% of SKUs with sales. Even more concentrated than the 28% reported by Strategy&.

• The cumulative curve rises above 100% and then falls back to 87.4%. That means the SKUs at the tail give back 12.6% of the profit earned up front. It’s a shape you will never see in a revenue ranking.

• Cutting the 34 loss-making SKUs lifts the contribution margin rate from 21.1% → 29.3% (+8.2pp) while keeping 82.4% of revenue. That’s a bigger jump than the “90–95% of revenue retained, 2–4pp margin improvement” McKinsey observed — because some of the loss-makers carry a large share of revenue.

• The bottom 50% of SKUs by revenue generate only 0.9% of revenue and 0.5% of contribution margin. This is the easy zone for rationalization decisions — but what you save here isn’t profit; it’s operating costs, inventory and your team’s time.

---

4-week action plan

---

Summary

1. If you stop at gross profit, the ranking doesn’t change. Its correlation with the revenue ranking is 0.99 — effectively the same list.

2. It flips at contribution margin. Correlation 0.58, an average shift of 3.1 places. The No. 2 product by revenue was last on contribution margin.

3. What flips it isn’t COGS — it’s ad spend. COGS rates differ by at most 16.5pp, while ad spend rates differ by 58.7pp.

4. Check three things before trusting the table. Fragmented product names (one product split into 19), ad spend allocation rules, and a 95% COGS match rate.

5. Cutting the 34 loss-making SKUs keeps 82.4% of revenue while lifting the margin by +8.2pp. But check each product’s role before deleting it.

Minimum settings to reproduce this report with your own data

---

Frequently asked questions

Q. What’s the difference between contribution margin and gross profit?

Gross profit is revenue minus product COGS only; contribution margin also subtracts variable costs such as fees, shipping and ad spend. In this sample there were 47.5pp between the 68.6% gross margin and the 21.1% contribution margin rate, and that gap is exactly where the ranking flipped.

Q. Should we cut every product with a low contribution margin rate?

No. Check its role first — if it’s an entry product that brings in new customers, or a cross-sell driver that lands in the cart alongside other items, you need to evaluate it including the follow-on purchases of the customers it brings in, even if its own contribution margin is low. The real phase-out candidates are low-revenue, low-margin products with no role.

Q. Can we start if our COGS data isn’t cleaned up?

It doesn’t have to be perfect. Even starting with average COGS rates by product category is enough to reveal the gap between revenue rank and profit rank. A realistic approach is to fill in actual COGS starting with the products that show the biggest gaps.

Q. Some channels show a negative contribution margin rate. Should we pull out of them?

Don’t judge by ratios alone. In small-revenue channels, the ratio easily explodes because of how ad spend is allocated (PART 05, Pitfall ②). First check the amount involved, and by what rule that ad spend was attached to the channel.

Q. What ROAS target should we set?

You can’t set just one. Break-even ROAS differs by product and by channel — in this sample it was 1.47–2.82x by product and 1.70–2.29x by channel. Calculate it yourself and set targets by product and by channel.

---

References: Strategy& (PwC) "Unlocking Hidden Value in Product Portfolios" · Deloitte Retail Profitability Study · McKinsey "The Power of Pricing" · McKinsey Assortment Optimization · Modern Retail, SKU rationalization at an oat-drink brand

The figures in this report are not real transaction data; they are illustrative values generated to reflect the distribution of Laplace benchmark metrics. They do not represent any specific customer, so please use them to understand trends rather than as absolute values. Figures from public research are cited with their source in each entry. Screen and setting names may differ slightly depending on the release.