The textbook definition is straightforward: revenue less variable costs. For a business selling one thing, that is usually enough to make the right cost decisions. For a business selling several products or services it is not, because a consolidated income statement blends the winners and the losers into a single number and never tells you which is which.
What follows is a practical process for running the analysis product by product, using both quantitative and qualitative inputs, and, just as important, the decisions the output does and does not support.
/ 01Contribution Margin Is Not Gross Margin
Gross profit subtracts everything sitting in cost of goods sold, fixed production cost included. Contribution margin subtracts only the cost that moves when volume moves. The two can sit far apart, and the size of that gap is itself a finding: a wide gap means variable selling and delivery costs are doing real damage below the gross profit line, which is exactly where most owners stop looking.
One definitional note before starting. Applied strictly, contribution margin subtracts variable cost only. In practice most operators run a traceable-cost version that also subtracts the direct, product-specific overhead a product would take with it if that product disappeared. That is the version used below. It is the more useful of the two for mix and pricing decisions, provided you stay clear-eyed that some of what you have subtracted is fixed.
/ 02How to Build the Analysis
Step 1Revenue
- Quantitative
- Compile every invoice issued for work delivered during the period, whether or not the customer has paid yet. Group the invoices into totals by product or service category. Use a trailing twelve months as the base period and cut it by quarter. Ninety days of paid invoices is a cash window, not an earnings window, and it will read a slow-collecting quarter as a bad quarter.
- Qualitative
- Record the average days to collect by category, and flag the customers who ran well past that average. Slow collection does not change the margin, but it changes what the margin is worth. A product that earns well and collects in 120 days is financing its own customers.
Step 2Cost of Goods Sold
- Quantitative
- For the same invoices, collect the corresponding vendor invoices for the materials and services required to deliver the work. Group them into the same product and service categories used in Step 1. Match each cost to the revenue it produced, not to the month the vendor happened to bill you.
- Qualitative
- Record the average time taken to receive materials or services from vendors, and when each vendor invoice was paid. Note any delivery delays outside the business's control.
Step 3Direct Operating Expenses
- Quantitative
- For the same invoices, identify the internal costs tied to delivering each product or service: sales commissions, delivery labour, implementation and support time, freight, payment processing. Group these into the same categories. Allocate only what is genuinely traceable, and leave everything shared unallocated at the bottom of the schedule. Spreading shared overhead across products on a revenue key is the fastest way to manufacture a loser that is not actually losing money.
- Qualitative
- Note the delivery delays that were inside the business's control, and which product lines they cluster around. Rework and overruns show up as cost long before they show up as a margin problem.
/ 03A Worked Example
Below is the income statement of a company selling two products, A and B. On a consolidated view it carries a strong 79% gross profit margin and a poor 1% EBITDA margin. Operating expenses are clearly the problem. What the statement cannot tell you is which product is causing it, because every operating cost has been pooled into three lines that span the whole business. This is where the contribution margin analysis earns its place.
Consolidated Income Statement
Exhibit 01All figures in thousands unless stated
| Consolidated | Total |
|---|---|
| Revenue | $ 1,400 |
| COGS | 300 |
| Gross Profit | 1,100 |
| Gross Profit % | 79% |
| General & Administrative Expense | 530 |
| Sales & Marketing Expense | 350 |
| Research & Development | 200 |
| EBITDA | $ 20 |
| EBITDA % | 1% |
Rebuilt by product, the same business looks entirely different.
Contribution Margin Analysis
Exhibit 02All figures in thousands unless stated
| Product / Service | A | + | B | = | AB |
|---|---|---|---|---|---|
| Revenue | $ 800 | $ 600 | $ 1,400 | ||
| COGS | 200 | 100 | 300 | ||
| Gross Profit | 600 | 500 | 1,100 | ||
| Gross Profit % | 75% | 83% | 79% | ||
| Direct General & Administrative Expense | 200 | 30 | 230 | ||
| Direct Sales & Marketing Expense | 100 | 50 | 150 | ||
| Contribution Margin | $ 300 | $ 420 | $ 720 | ||
| Contribution Margin % | 38% | 70% | 51% | ||
| General & Administrative Expense | 300 | ||||
| Sales & Marketing Expense | 200 | ||||
| Research & Development | 200 | ||||
| EBITDA | $ 20 | ||||
| EBITDA % | 1% |
Product A is the larger line, at 57% of revenue, and carries a respectable 75% gross margin. Once its direct costs are charged against it, the contribution margin falls to 38%. Product B, on less than half the revenue, converts 70% of its sales into contribution and throws off more contribution dollars than A does: $420K against $300K. Product B is carrying the overhead of this business. Product A is spending most of what it earns on the way to the customer.
Note the gap between the 79% consolidated gross margin and the 51% consolidated contribution margin. That 28-point spread is the part of the business the income statement was hiding, and almost all of it sits in Product A.
The right responses, in order: reprice Product A, take cost out of delivering it, or shift mix toward Product B.
/ 04What the Analysis Does Not Say
It does not say drop Product A.
Product A still contributes $300K against $700K of company costs that would not disappear with it. Walk away from Product A tomorrow and, unless the cost of serving it walks away too, EBITDA goes from positive $20K to negative $280K. The test for exiting a product line is never its margin ranking. It is the contribution you would lose measured against the cost you could genuinely avoid, and most overhead is far less avoidable than it looks on a schedule.
The same caution applies to shifting mix. Selling more of Product B only helps if the demand is there and the capacity freed from A can actually be redeployed to B. Otherwise the analysis has identified a pricing problem, not a portfolio problem.
/ 05Where These Analyses Go Wrong
Three failures account for most of them.
- Cost misclassification. A cost is variable only if it genuinely moves with volume. Semi-variable costs, utilities and maintenance among them, get forced into one bucket or the other and quietly distort every number downstream.
- Over-allocation. Shared cost gets spread on a revenue key because leaving it unallocated feels incomplete. It is not incomplete. It is correct.
- The wrong period. Cash-basis windows, a single quarter in a seasonal business, and cost matched to the wrong revenue will each produce a confident answer that is wrong.
Run the analysis on the same definitions every period, and reconcile the totals back to the income statement. If product-level revenue and cost do not tie to the statutory numbers, the analysis will not survive the first person who checks it.
Written by Alex Balca, Scepter Capital.
All Insights →