Daniel has a problem most business owners would initially be happy to have.
One of his products is selling extremely well.
It generates €18,000 in monthly revenue.
Customers like it.
Returns are low.
The gross margin looks respectable.
And every month, Daniel’s sales report puts the product near the top of the list.
Then his operations manager asks an uncomfortable question:
„Have you ever calculated how much time this product consumes?“
Daniel has not.
He knows what the product costs to buy.
He knows its selling price.
He knows the payment fees.
He knows roughly what shipping costs.
What he has never measured is everything that happens between receiving an order and finally being finished with it.
So for one month, the company starts counting.
The result changes the way Daniel looks at his bestseller.
💰 The Product Looks Excellent on the Sales Report
Daniel sells a customized home-office accessory.
Average selling price:
€120
Monthly orders:
150
Revenue:
150 × €120 = €18,000
The direct product cost is €58.
Packaging costs €4.
Payment fees average €3.
Average outbound shipping paid by the company is €9.
That gives Daniel:
| Per order | Amount |
|---|---|
| Selling price | €120 |
| Product cost | -€58 |
| Packaging | -€4 |
| Payment fee | -€3 |
| Shipping | -€9 |
| Remaining contribution | €46 |
At 150 orders:
150 × €46 = €6,900
That looks good.
Daniel is not making the classic mistake of confusing revenue with profit. He already knows that €18,000 of sales is not €18,000 earned.
But the €6,900 still gives him confidence.
The product appears to be contributing a meaningful amount toward salaries, rent, software, marketing and profit.
There is just one problem.
The calculation treats every order as though the work ends when the shipping label is printed.
It does not.
⏱️ Daniel Measures What Happens After „Buy Now“
The product is customizable.
Customers choose a size, finish and several configuration options.
That creates work.
The team records the average time spent on one order.
Before shipment
🟢 Order review: 4 minutes
🟢 Customization check: 6 minutes
🟢 Preparation and packing: 11 minutes
🟢 Shipping administration: 4 minutes
Total:
25 minutes
That still seems manageable.
Then Daniel includes the work that is easier to forget.
Some customers send questions before ordering.
Others change specifications after ordering.
A small number enter incomplete information.
Some shipments generate delivery questions.
A few products are returned.
Not every order creates every task, so Daniel calculates the average across all 150 orders.
Hidden service workload per order
🔵 Pre-sale questions: 5 minutes average
🔵 Changes and corrections: 4 minutes average
🔵 Post-sale support: 6 minutes average
🔵 Returns and problem handling: 3 minutes average
Additional average workload:
18 minutes
The real average becomes:
25 + 18 = 43 minutes per order
Across 150 orders:
150 × 43 minutes = 6,450 minutes
or:
107.5 hours per month
Daniel’s bestseller is consuming more than 107 working hours every month.
Now the €6,900 looks different.
🧮 What Is an Hour of Capacity Actually Worth?
Daniel could simply divide:
€6,900 ÷ 107.5 hours ≈ €64.19 per hour
But even this number needs interpretation.
The €6,900 is not pure profit. It still has to help pay fixed operating costs.
And the 107.5 hours are not free.
Employees are being paid during those hours.
Suppose the fully loaded internal labor cost for the people handling these tasks averages:
€28 per hour
Then the labor consumed is approximately:
107.5 × €28 = €3,010
After including this operational labor:
€6,900 – €3,010 = €3,890
The product still contributes positively.
So Daniel should not discontinue it simply because hidden labor exists.
But he has discovered something much more important:
The scarce resource may not be the product. It may be the company’s time.
That matters when the business has spare capacity.
It matters much more when the team is already full.
📦 Compare It With a Product That Barely Gets Attention
Daniel sells another product.
It generates only:
€11,000 monthly revenue
So it looks weaker in his sales dashboard.
The product is standardized.
No customization.
Very few questions.
Simple packaging.
Average selling price:
€110
Monthly orders:
100
Direct costs leave:
€38 contribution per order
Monthly contribution before operational labor:
100 × €38 = €3,800
At first glance:
| Product | Bestseller A | Product B |
|---|---|---|
| Revenue | €18,000 | €11,000 |
| Contribution before labor | €6,900 | €3,800 |
| Orders | 150 | 100 |
| Dashboard winner | 🏆 A | — |
Daniel would naturally protect Product A.
Then he measures Product B’s workload.
Average total handling time:
12 minutes per order
Monthly workload:
100 × 12 = 1,200 minutes
or:
20 hours
At €28 internal labor cost:
20 × €28 = €560
Contribution after this labor:
€3,800 – €560 = €3,240
Now compare the two again.
| Metric | Product A | Product B |
|---|---|---|
| Revenue | €18,000 | €11,000 |
| Contribution before measured labor | €6,900 | €3,800 |
| Operational hours | 107.5 h | 20 h |
| Measured labor cost | €3,010 | €560 |
| Contribution after measured labor | €3,890 | €3,240 |
Product A still produces more money.
But only:
€650 more
while consuming:
87.5 additional hours
That is a radically different business picture.
⚙️ Revenue per Order Was Hiding the Constraint
Daniel calculates another measure.
How much contribution after measured operational labor does each product generate per hour of the operational capacity it consumes?
For Product A:
€3,890 ÷ 107.5 ≈ €36.19 per operational hour
For Product B:
€3,240 ÷ 20 = €162 per operational hour
This does not mean Product B has a magical 162-euro hourly profit margin.
Fixed costs, marketing, management time, inventory financing and other expenses still exist.
It is a capacity comparison.
And that comparison tells Daniel something his revenue report never showed.
If his warehouse and service team have plenty of unused time, Product A’s workload may be acceptable.
If the team is already overloaded, every hour allocated to Product A has an opportunity cost.
That hour cannot simultaneously be used for something else.
🚧 The Problem Appears When Capacity Becomes Full
Imagine Daniel’s current team can comfortably provide:
500 operational hours per month
Existing products already consume:
450 hours
That leaves:
50 hours spare
Now a marketing campaign doubles demand for Product A.
Another 150 orders arrive.
They require approximately:
107.5 additional hours
But Daniel only has 50 hours available.
He is short:
57.5 hours
Something has to happen.
He can:
🔧 reduce the work per order,
👤 hire additional staff,
⏳ create longer delivery times,
❌ reject some demand,
💶 raise the price,
or
📉 allow service quality elsewhere to deteriorate.
This is the moment when a profitable product can create a business problem.
Not because each sale loses money.
Because the sales consume more of a scarce resource than the company can supply efficiently.
👤 Hiring One More Person Changes the Economics Again
Suppose Daniel hires additional support.
The new employee costs the business approximately:
€4,200 per month
and provides around:
140 realistically productive operational hours after meetings, administration and other unavoidable time.
The extra 150 Product A orders generate another:
€6,900 contribution before measured operational labor
On paper, hiring seems obvious.
But the employee is not a perfectly divisible resource.
Daniel cannot buy exactly 57.5 hours of permanent monthly capacity if the business needs a full employee.
The decision is therefore not:
„Are the extra orders profitable?“
It becomes:
„Does the additional demand justify the next block of capacity we must purchase?“
That distinction is crucial.
Businesses often grow in steps.
One warehouse works until it is full.
Then another space is needed.
Three employees can handle the workload until a fourth is required.
One customer-service system works until complexity demands a more expensive platform.
A van handles deliveries until another vehicle and driver become necessary.
Costs therefore do not always rise smoothly with revenue.
Sometimes they jump.
📈 Daniel Tests a Price Increase Before Hiring
Instead of immediately adding capacity, Daniel runs another calculation.
What happens if Product A rises from:
€120 to €135?
Assume direct costs remain €74 per order:
- €58 product,
- €4 packaging,
- €3 payment cost for simplicity,
- €9 shipping.
Contribution before measured labor becomes:
€135 – €74 = €61
instead of:
€46
That is an increase of:
€15 per order
or roughly:
32.6% more contribution per order
Now assume the higher price reduces monthly demand from 150 orders to 125.
Revenue becomes:
125 × €135 = €16,875
Revenue has fallen from €18,000.
A sales dashboard might show that as bad news.
But contribution before measured labor becomes:
125 × €61 = €7,625
That is higher than the original €6,900.
And workload falls too.
At 43 minutes per order:
125 × 43 = 5,375 minutes
or about:
89.6 hours
Daniel now has:
- lower revenue,
- fewer orders,
- less operational work,
- and more contribution before measured labor.
🔎 This is the kind of result that revenue growth alone can completely hide.
💡 Example: The „Successful“ Promotion
Daniel now understands why one of last year’s promotions felt strangely exhausting.
The company offered a 15% discount.
The product fell from €120 to:
€102
Orders jumped from 150 to:
220
Everyone celebrated the sales volume.
But let’s examine it.
Using the same illustrative €74 direct cost:
Contribution per discounted order:
€102 – €74 = €28
At 220 orders:
220 × €28 = €6,160
Without the promotion:
150 × €46 = €6,900
So the promotion produced:
📈 46.7% more orders
but
📉 €740 less contribution before measured operational labor
And workload increased from approximately:
107.5 hours
to:
157.7 hours
The company processed roughly 50 additional hours of work while generating less contribution.
The promotion was a spectacular success if Daniel measured orders.
It was much less impressive if he measured economics.
🧠 The Right Question Is Not „Which Product Sells Best?“
Daniel no longer ranks products using one number.
Revenue tells him something useful.
Contribution tells him something else.
Operational workload adds another dimension.
Capacity requirements add another.
Customer acquisition can change the picture again.
So can repeat purchases.
A product with mediocre first-order economics may be valuable if it reliably creates profitable long-term customers.
A labor-intensive product may be strategically important because it differentiates the company.
A low-volume product may deserve to stay because customers routinely purchase higher-margin products alongside it.
There is no universal metric that automatically identifies which product to remove.
But there is a much better question:
What does this product contribute relative to the scarce resources it consumes?
For Daniel, that question reveals something his bestseller list never could.
The product at the top of the revenue chart may still deserve to stay.
It may even deserve more investment.
But first, Daniel needs to know whether he is selling a highly profitable product —
or simply selling a very large amount of work.
📊 Daniel Builds a Product Scorecard Instead of a Bestseller List
Daniel does not want another complicated dashboard that nobody checks after two weeks.
He wants a system that answers a practical question:
Which products deserve more of the company’s limited money, time and attention?
So he starts with six factors.
| Factor | What Daniel measures |
|---|---|
| 💰 Contribution | Money left after directly attributable costs |
| ⏱️ Workload | Operational time consumed |
| 📦 Inventory | Cash and storage tied up |
| ↩️ Returns | Cost and workload created after the sale |
| 🎧 Support | Questions, changes and problem cases |
| 🔁 Customer value | Repeat purchases and additional sales |
Revenue remains on the dashboard.
It simply loses its position as the automatic winner.
📦 Inventory Can Make an Easy Product Expensive
Product B looked excellent because it required only 20 operational hours per month.
But Daniel discovers another weakness.
He keeps approximately:
€45,000 of Product B inventory
on hand.
Product A requires only:
€16,000
because suppliers can replenish it quickly.
Now Product B is consuming less labor but considerably more working capital.
That €45,000 cannot simultaneously pay suppliers, fund marketing or support another product launch.
There is also storage.
Suppose Product B occupies:
18 pallet positions
while Product A uses:
6
If warehouse capacity is abundant, the difference may barely matter.
If Daniel is about to rent additional space, those pallet positions suddenly have economic value.
This is why a product cannot be judged only by what happens when it sells.
Daniel also needs to ask:
What does it cost us to keep this product ready to sell?
↩️ Returns Can Reverse an Attractive Margin
A third product, Product C, appears fantastic at first.
Selling price:
€160
Direct costs:
€85
Initial contribution:
€75
That is much stronger than Product A.
But Product C has a return rate of:
18%
Product A’s return rate is only:
4%
Daniel looks beyond the refund itself.
A returned item can create:
🔸 return shipping,
🔸 inspection,
🔸 repackaging,
🔸 customer-service time,
🔸 payment adjustments,
🔸 damaged packaging,
🔸 discounted resale,
🔸 or occasionally an item that cannot economically be resold.
Suppose the average additional cost of processing a Product C return is €24, excluding the refunded purchase price.
Across 100 orders:
18 returns × €24 = €432
But that still understates the impact if returned inventory loses value.
Imagine six of those returned units can only be resold for €30 less than normal.
Additional loss:
6 × €30 = €180
Now the return-related burden is:
€432 + €180 = €612 per 100 orders
The original €75 contribution per successful-looking order did not reveal any of this.
🎧 Ten Support Tickets Are Not the Same as Ten Orders
Daniel then analyzes customer support.
Product A generates roughly:
38 support interactions per 100 orders
Product B generates:
7
Product C generates:
54
But even ticket counts are incomplete.
A simple delivery-status question might take three minutes.
A technical compatibility problem might consume 25 minutes and require escalation.
So Daniel measures both frequency and time.
Example support workload per 100 orders
| Product | Support cases | Avg. time per case | Total support time |
|---|---|---|---|
| Product A | 38 | 11 min | 418 min |
| Product B | 7 | 6 min | 42 min |
| Product C | 54 | 16 min | 864 min |
Product C consumes:
14.4 support hours per 100 orders
Product B consumes:
0.7 hours
That difference becomes particularly important when the same support team serves the entire business.
Product C is not merely consuming its own margin.
It can create slower response times for customers who bought other products.
That is a cost that rarely appears beside the product in an ordinary sales report.
🔁 But a Difficult Product Can Still Be Worth Keeping
Daniel is careful not to create another simplistic rule.
High support does not automatically mean:
Delete the product.
Product C might attract unusually valuable customers.
Suppose 100 first-time Product C buyers generate only modest profit initially.
But within twelve months:
42 of them purchase again
and many move into a highly profitable product category.
Meanwhile Product B attracts bargain hunters who rarely return.
Suddenly Product C’s workload may be an acquisition cost rather than pointless inefficiency.
Daniel therefore checks customer behavior.
He asks:
What happens after this sale?
Not:
What could theoretically happen over a customer’s lifetime?
There is an important difference.
He uses actual cohorts where possible.
If customers who first bought Product C have historically produced €210 of additional contribution during the following twelve months, that is useful evidence.
If someone simply assumes:
„Our average customer stays five years, so every new customer is worth €2,000“
the number can become fantasy disguised as precision.
🛒 Cross-Selling Can Rescue a Weak Standalone Product
Product D provides another lesson.
On its own, it is unimpressive.
Contribution after directly attributable costs:
€17 per order
Daniel considers removing it.
Then he checks baskets containing Product D.
He discovers that customers frequently buy two accessories with it.
Average basket without Product D:
€92
Average basket containing Product D:
€148
More importantly, the accessories have strong margins.
Across 500 Product D orders, the associated accessories generate an additional:
€9,500 contribution
That changes the decision.
Product D may be weak as an isolated SKU but valuable as part of a basket.
This is why Daniel does not automatically delete everything below an arbitrary margin threshold.
Products can play different roles:
🟢 Profit generator
🟢 Customer acquisition product
🟢 Cross-sell driver
🟢 Retention product
🟢 Strategic differentiator
🟡 Necessary range filler
A product should ideally have a reason for consuming resources.
„We’ve always sold it“ is not a particularly strong one.
🚦 Daniel Creates a Green, Yellow and Red Test
He now gives every important product a simple operational assessment.
🟢 GREEN — protect or expand
A product tends toward green when it:
- produces healthy contribution,
- uses capacity efficiently,
- has manageable returns,
- requires little exceptional support,
- does not tie up disproportionate inventory,
- and/or creates demonstrable value elsewhere in the customer relationship.
Green does not mean perfect.
It means the product earns its place in the business.
🟡 YELLOW — fix before scaling
Yellow products may be profitable but contain a structural weakness.
Examples include:
⚠️ strong sales but excessive customization,
⚠️ good margin but high return rates,
⚠️ attractive demand but poor supplier terms,
⚠️ good customer acquisition but weak first-order economics,
⚠️ profitable orders that consume too much scarce staff capacity.
Daniel does not necessarily remove yellow products.
He asks whether the weakness can be redesigned.
🔴 RED — challenge its reason for existing
A product moves toward red when several problems combine.
Low contribution.
High workload.
High returns.
Slow inventory.
Frequent support.
Weak repeat purchasing.
Little cross-selling.
No strategic purpose.
Daniel’s question then becomes:
If we did not already sell this product, would we deliberately introduce it today?
That question is uncomfortable.
It is also useful.
🔧 A Yellow Product Can Often Be Repaired
Product A is Daniel’s first yellow candidate.
It sells.
Customers like it.
The economics are not disastrous.
The problem is the 43 minutes of average workload.
So Daniel breaks those 43 minutes apart.
| Activity | Avg. minutes/order | Can it be reduced? |
|---|---|---|
| Order review | 4 | 🟢 Probably |
| Customization check | 6 | 🟢 Probably |
| Preparation & packing | 11 | 🟡 Partly |
| Shipping administration | 4 | 🟢 Yes |
| Pre-sale questions | 5 | 🟢 Probably |
| Changes & corrections | 4 | 🟢 Strong opportunity |
| Post-sale support | 6 | 🟡 Investigate |
| Returns/problems | 3 | 🟡 Investigate |
| Total | 43 |
Daniel realizes the product itself may not be the problem.
The process is.
Customers enter customization information in a free-text field.
Employees manually interpret it.
When something is unclear, they send an email.
Customers reply.
Orders wait.
Sometimes the wrong configuration is selected and corrected later.
Daniel replaces the free-text process with structured choices, validation and a visual confirmation before checkout.
Suppose that removes:
6 minutes per average order
Shipping software integration saves another:
3 minutes
Better pre-purchase information saves:
2 minutes of average questions
Total saving:
11 minutes per order
New workload:
43 – 11 = 32 minutes
At 150 orders:
Old workload:
107.5 hours
New workload:
80 hours
Monthly capacity recovered:
27.5 hours
At €28 measured labor cost, that represents:
€770 of monthly labor capacity
without selling one additional unit.
Daniel has effectively created capacity through process design.
💶 Then He Tests Whether Customers Should Pay for Complexity
Not every customer chooses the difficult configuration.
Daniel discovers:
70% select standard options.
30% create most of the customization work.
Until now, everyone paid the same price.
That means easy customers were indirectly subsidizing complicated orders.
Daniel introduces a €20 customization charge for the labor-intensive option.
If 45 monthly orders use it:
45 × €20 = €900 additional revenue
The fee may also discourage customers who do not genuinely value the customization.
That can be beneficial.
A business does not always need to make every variation equally attractive.
Sometimes price should communicate:
This option consumes additional resources.
🧮 The Most Valuable Product Depends on What Is Scarce
Daniel finally understands why product profitability cannot be reduced to one universal ranking.
If cash is scarce, inventory efficiency may dominate.
If warehouse space is scarce, cubic meters or pallet positions matter.
If the support team is overloaded, support minutes matter.
If production equipment is full, machine hours matter.
If skilled specialists are the constraint, their hours may be the most valuable resource in the company.
Consider two products:
| Product X | Product Y | |
|---|---|---|
| Contribution per unit | €80 | €45 |
| Scarce machine time | 40 min | 10 min |
| Contribution per machine hour | €120 | €270 |
If machine capacity is the bottleneck, Product Y may deserve priority despite producing much less contribution per unit.
But if the machine sits idle half the week, that calculation becomes far less important.
A constraint matters because it is actually constrained.
📋 The Monthly Product Review Daniel Actually Uses
Daniel reduces everything to a review he can repeat.
For each major product or product family, he tracks:
| Question | Measure |
|---|---|
| 💰 Does it generate money? | Contribution € and % |
| ⏱️ Does it consume scarce time? | Minutes/order and contribution/capacity hour |
| 📦 Does it trap cash? | Inventory value and turnover |
| ↩️ Does it come back? | Return rate and return cost |
| 🎧 Does it create work later? | Support cases and minutes |
| 🛒 Does it improve the basket? | Cross-sell contribution |
| 🔁 Does it create better customers? | Observed repeat contribution |
| ⚙️ Can the process improve? | Avoidable minutes/costs |
| 🚦 What action follows? | Green / Yellow / Red |
He does not need perfect accounting allocation for every paper clip.
The purpose is to identify material differences that change decisions.
If two products differ by 80 operational hours per month, pretending that workload does not exist because it is difficult to allocate precisely is worse than using a reasonable measured estimate.
📉 Sometimes Lower Revenue Is the Better Result
Six months later, Daniel’s business looks slightly strange from the outside.
One low-quality promotion has disappeared.
A complex customization option costs more.
Two weak products have been removed.
Several repetitive tasks have been automated.
One high-revenue product sells fewer units than before.
Total revenue has barely grown.
That could look disappointing on a chart built around sales alone.
But:
🟢 contribution is higher,
🟢 overtime has fallen,
🟢 support responds faster,
🟢 less cash is sitting in slow inventory,
🟢 warehouse capacity is healthier,
🟢 and the company can accept more of the orders it actually wants.
Daniel has stopped asking his products to win a revenue contest.
He asks each one to justify the resources it consumes.
That produces a very different definition of a bestseller.
The best product is not necessarily the one that creates the largest number on the sales report.
It is the one that creates enough economic value for the money, time, space, attention and capacity the business has to give it.