If your business is winning plenty of work but still not making enough money, you probably do not have a sales problem.

You have a margin leakage problem somewhere between winning the work and completing it.

The business is successfully creating revenue.

Then something happens to the profit.

Perhaps the work was underpriced.

Perhaps sales discounted it.

Perhaps labour took longer than estimated.

Perhaps materials cost more than expected.

Perhaps the customer added another 20 per cent of work without paying another 20 per cent.

Perhaps rework swallowed the margin.

Perhaps scheduling is poor.

Perhaps employees are busy but not productive.

Perhaps overhead has grown faster than gross profit.

Perhaps you are winning plenty of work, but the wrong work.

The answer is not automatically to sell more.

First find out where the profit disappears.

Start by separating revenue from profit

This sounds painfully basic.

It also causes enormous confusion.

You can have:

Record sales.

A full order book.

Employees working flat out.

Customers queuing.

And disappointing profit.

Revenue tells you how much you sold.

It does not tell you how much money remained after delivering what you sold.

The basic journey looks something like:

Sales revenue

minus:

The direct costs of delivering those sales

leaves:

Gross profit or contribution

Then you still have to cover:

Overheads

before arriving at:

Operating or net profit, depending on exactly how your accounts are structured.

The terminology and accounting treatment will differ between businesses, so your accountant should help establish the numbers that matter for your company.

But the management question is universal:

How much of each pound we sell actually survives?

If you do not know that, winning more work can make you busier without making you meaningfully richer.

First check whether you mean profit or cash

Before going any further, establish what problem you actually have.

Sometimes owners say:

“We're making no money.”

What they mean is:

“There's no bloody money in the bank.”

Those are related but different problems.

Your company can be profitable but cash-poor because:

Customers pay slowly.

Stock consumes cash.

Work in progress has not yet been invoiced.

Growth requires expenditure before revenue arrives.

Tax is due.

Large deposits have been paid.

Equally, the bank account can temporarily look healthy while the underlying work is barely profitable.

This article is primarily about profitability.

If the accounts show healthy margins but cash is disappearing, investigate working capital and cash flow separately.

Do not try to repair one problem using the tools for the other.

Follow the profit through the job

Imagine you quote a project for:

£100,000

You expect:

£45,000 labour and materials.

£55,000 gross profit before overhead allocation.

The job looks perfectly respectable.

Then it finishes.

Actual direct costs were:

£61,000.

The customer still pays £100,000.

Nothing obviously catastrophic happened.

The project was delivered.

The customer is happy.

Revenue appears exactly as forecast.

But £16,000 of expected gross profit disappeared.

Your problem is not sales.

Your problem is somewhere inside the £16,000 variance.

Find it.

That is the central discipline throughout this article:

Compare what you thought the work would cost with what it actually cost.

Problem 1: You are underpricing the work before it begins

Sometimes the margin never existed.

The quote looked profitable because the costing assumptions were wrong.

Perhaps you underestimated:

Labour hours.

Materials.

Subcontractors.

Travel.

Waste.

Delivery.

Management time.

Equipment.

Commission.

Installation.

Warranty.

Compliance costs.

Setup.

Project administration.

Some businesses price from experience.

That can work extremely well when the experience is current and accurate.

It becomes dangerous when the number is essentially:

“We normally charge about £12,000 for those.”

Why?

What does it cost?

What margin does it create?

Has the cost changed?

Nobody quite knows.

British Business Bank guidance recommends understanding fixed and variable costs and contribution margin when assessing profitability, because sales volume on its own does not tell you whether individual products or services are contributing enough to cover the wider cost base.

You do not need every salesperson to become a management accountant.

But somebody needs to understand the economics behind the price.

Margin and markup are not the same thing

This catches people out regularly.

Suppose something costs you £70.

You want a 30 per cent margin.

If you simply add 30 per cent to the cost, you get:

£91.

But your profit is £21 on £91 of sales.

That is only around a 23 per cent margin.

To achieve a 30 per cent margin on a £70 cost, you need to sell for £100.

Because £30 profit is 30 per cent of the £100 selling price.

Markup and margin answer different questions.

Markup measures profit relative to cost.

Margin measures profit relative to selling price.

If different people inside the business use those words interchangeably, your quotations can look considerably healthier than they actually are.

Get the language consistent.

Problem 2: You keep discounting the profit away

Sales wins the work.

Everybody celebrates.

Then you discover what was required to win it.

A little discount.

Then another.

Then free delivery.

Then longer payment terms.

Then one extra item included.

Each concession feels manageable individually.

Together, they can dismantle the economics.

Consider work with a 30 per cent contribution margin.

A 10 per cent reduction in selling price does not reduce your contribution by 10 per cent.

If costs remain unchanged, contribution drops from 30 to 20.

You now need 50 per cent more volume to generate the same total contribution as before.

ICAEW uses exactly this type of cost-volume-profit example to show why apparently modest price changes can have disproportionately large effects on profitability.

This is why:

“We'll make it up on volume.”

deserves maths.

Not optimism.

Give people discount authority, not unlimited discretion

If salespeople can discount, define the boundaries.

For example:

What discount can they approve themselves?

When does management approval become necessary?

What minimum margin must remain?

What concessions can be exchanged for something useful?

Could a discount require:

Larger volume?

A longer commitment?

Faster payment?

Reduced service?

Standard specification?

A deposit?

Do not train customers that the quoted price is merely the opening move in a game where persistent negotiation always wins.

And do not reward salespeople purely for revenue if the company ultimately needs profitable work.

People tend to optimise what you measure.

Problem 3: Scope creep is eating the job alive

The project was priced correctly.

Then the project changed.

One more revision.

One additional visit.

A little extra reporting.

Another meeting.

A different specification.

The client asks:

“While you're here...”

Nobody wants to be difficult.

So the business absorbs it.

Repeatedly.

Three months later, a profitable job has become mediocre.

The customer might not even realise.

They asked.

You said yes.

The problem is not necessarily the customer.

It may be the absence of a system for recognising when normal service becomes additional scope.

Good commercial control asks:

What was included?

What has changed?

What does the change cost?

Does it affect programme?

Does it need approval?

Does it need charging?

You can be flexible without making flexibility free.

Small extras become large leaks

Imagine five employees each provide two uncharged additional hours every week.

That is ten hours.

Across 48 working weeks:

480 hours.

Then add management involvement.

Travel.

Materials.

Administration.

The company may be giving away hundreds of productive hours without anything appearing on the sales report.

This is why profit leakage can be difficult to see.

Nobody writes:

“Lost £18,000 through lots of small favours.”

It is simply hidden inside labour utilisation and gross-margin erosion.

Problem 4: Labour is taking longer than estimated

For many service, construction, technical and project businesses, labour variance is one of the biggest places margin disappears.

You quoted:

40 hours.

Actual:

Why?

Perhaps the estimate was wrong.

Perhaps the employee was inexperienced.

Perhaps access was poor.

Perhaps the customer delayed things.

Perhaps preparation was inadequate.

Perhaps information was missing.

Perhaps equipment failed.

Perhaps the team spent six hours waiting.

Perhaps work had to be repeated.

Do not immediately blame the employee.

The interesting question is:

Why did actual hours differ from expected hours?

Then find out whether the cause is:

Estimating.

Training.

Planning.

Scheduling.

Supervision.

Customer behaviour.

Process.

Or genuinely poor individual performance.

One number can reveal the problem.

It cannot diagnose it for you.

Track estimated hours against actual hours

You do not need to measure employees like laboratory rats.

But if labour is a major cost and you never compare estimated labour with actual labour, how will you improve pricing?

Your estimating team needs feedback from delivery.

Otherwise:

Estimators keep making the same assumptions.

Operations keeps discovering reality.

Finance reports the damage six weeks later.

Nobody closes the loop.

A mature system asks:

What did we estimate?

What actually happened?

Why?

What should we change next time?

That learning gradually improves both pricing and operational performance.

Problem 5: Rework is destroying margin

Rework is particularly unpleasant because the customer may still receive exactly what they originally bought.

You simply delivered it twice.

Wrong dimensions.

Poor workmanship.

Incorrect information.

Missed requirement.

Damaged material.

Incomplete specification.

Miscommunication.

The second attempt often generates no additional revenue.

Just additional cost.

Rework also creates secondary damage:

Scheduling disruption.

Overtime.

Expedited materials.

Management time.

Customer frustration.

Delayed jobs elsewhere.

Teams staying late.

One error can therefore cost considerably more than the obvious replacement item.

Track recurring causes.

If the same error happens repeatedly, stop treating each occurrence as bad luck.

You have a process problem.

Problem 6: Purchasing and material costs have moved

You quoted using:

Old supplier pricing.

Historic material costs.

A rough allowance.

Then reality arrived.

Material inflation.

Delivery surcharges.

Minimum order charges.

Different specification.

Supplier substitution.

Wastage.

Rush delivery.

Your quote remains fixed.

Your cost doesn't.

That gap comes straight out of margin unless your contract allows the increase to be recovered.

Regularly update costing data.

Especially where material inputs represent a significant proportion of job value.

A perfect estimating formula using 18-month-old costs still produces a bad quote.

Problem 7: Your scheduling creates expensive idle time

A company can be incredibly busy and still operate inefficiently.

Employees travel between jobs unnecessarily.

Jobs start before required information is available.

Teams wait for other trades.

Equipment sits unused.

Skilled people perform low-value work because nobody else is available.

One department is overloaded while another has spare capacity.

Overtime appears because ordinary working hours were badly coordinated.

This is operational waste.

Not every minute needs monetising.

People are human.

Work contains natural variation.

But poor planning repeatedly converts paid hours into little or no customer value.

That directly affects margin.

Utilisation matters, but do not abuse it

Utilisation measures can be useful.

They can also become bloody ridiculous.

A consultant being 75 per cent billable might be sensible if the remaining time is needed for:

Training.

Internal meetings.

Business development.

Administration.

Thinking.

A technical employee may require setup, maintenance or travel time.

Trying to achieve 100 per cent productive utilisation can simply move necessary work somewhere else or encourage people to misrecord time.

The point is not maximum utilisation.

The point is understanding where paid capacity goes.

If payroll increases 30 per cent while productive output rises 5 per cent, investigate.

Problem 8: You are winning work that does not fit your capacity

Imagine your team is excellent at Project Type A.

It runs smoothly.

Good margin.

Predictable labour.

Then sales wins Project Type B.

Similar revenue.

Much greater complexity.

Different skill requirements.

More meetings.

Longer setup.

Specialist subcontractors.

The company can technically deliver it.

But badly.

The problem isn't necessarily pricing alone.

The work does not fit the operating system.

Businesses often discover this while growing.

Sales expands faster than operational capability.

The order book looks fantastic.

The company then spends the next six months forcing unsuitable work through a system never designed to deliver it.

Revenue grows.

Chaos grows faster.

Growth can expose hidden inefficiency

At £1 million turnover, the owner can personally patch problems.

At £3 million, there are too many.

More work now requires:

More coordination.

More supervisors.

More administration.

More project management.

More systems.

More finance support.

More recruitment.

More communication.

The relationship between revenue and cost is not perfectly linear.

Sometimes you cross a threshold where the next £1 million of turnover requires a completely different operating structure.

The business then experiences a frustrating period where sales rise but profit does not.

That does not necessarily mean growth was a mistake.

It may mean the company has entered an investment step.

But you need to understand whether the additional cost is:

Temporary infrastructure for future scale.

Or permanent inefficiency you have accidentally normalised.

Those are very different things.

Problem 9: Overhead has grown faster than gross profit

Perhaps job margins are actually fine.

The damage happens afterwards.

You have added:

Managers.

Software.

Premises.

Vehicles.

Marketing.

HR.

Finance.

Administration.

Subscriptions.

Consultants.

Equipment finance.

Each decision had a justification.

Then the company discovers that gross profit did not grow enough to carry the enlarged overhead base.

Contribution and cost-volume-profit analysis exist partly to answer this question: how much contribution from sales is required to cover fixed costs before profit appears? ACCA's guidance explains that contribution is what remains after variable costs and that this contribution must first cover fixed costs before the business generates profit.

This is why turnover targets in isolation can be fairly meaningless.

You need enough contribution.

Do you know your break-even point?

You should.

Roughly, at least.

How much revenue does the business need each month to cover:

Direct costs.

Payroll.

Premises.

Finance.

Software.

Vehicles.

Management.

Administration.

Everything else required to operate.

The precise accounting calculation depends on your cost structure and sales mix.

Your accountant can help you establish it properly.

But an owner should understand the broad answer.

If the company needs £400,000 monthly revenue to break even at its current margin and you average £390,000, the problem looks very different from a company breaking even at £250,000.

One needs growth or margin improvement immediately.

The other has substantial room to investigate why profit isn't translating into cash or where accounting treatment differs.

Know the number.

Problem 10: You are selling the wrong mix of work

Not all revenue is equal.

Suppose you sell:

High-margin maintenance.

Medium-margin installations.

Low-margin large projects.

Revenue shifts heavily towards large projects.

Turnover increases.

Average margin falls.

Nothing necessarily went wrong operationally.

Your sales mix changed.

This is particularly important when people are rewarded on revenue rather than contribution.

A salesperson may rationally prefer:

One £500,000 contract at 12 per cent margin

over:

Five £100,000 contracts at 30 per cent margin

because their commission or target is based purely on revenue.

The business may prefer the opposite.

Align commercial incentives with what the company actually needs.

Problem 11: Your biggest customers may have the worst economics

Article #65 covered this in more detail.

A customer can generate substantial sales and absorb an even more substantial amount of resource.

Discounting.

Rework.

Senior attention.

Special reporting.

Long payment terms.

Bespoke processes.

Rush demands.

Customer profitability matters because aggregate company figures hide variation.

One group of customers may be generating excellent contribution while another quietly consumes it.

Do not only analyse:

What do we sell?

Also analyse:

Who do we sell it to?

Problem 12: Sales and operations are optimising different things

This creates one of my favourite business arguments.

Sales says:

“We won it.”

Operations says:

“Why the fuck did you sell that?”

Both departments may be performing against the measures given to them.

Sales is rewarded for bookings.

Operations is judged on delivery.

So sales promises:

Shorter lead times.

Custom requirements.

Special conditions.

Lower prices.

Then operations inherits the complexity.

You need commercial discipline before the contract is signed.

For larger or unusual work, involve delivery early enough to answer:

Can we actually deliver this?

At what cost?

With what resource?

What are the risks?

What assumptions are inside the price?

What happens if those assumptions are wrong?

The handover from sales to operations should not feel like somebody throwing a live grenade over a wall.

Problem 13: Variations are happening but not being captured

Particularly in project businesses, work changes.

That is normal.

The problem is when the commercial record doesn't.

Customer requests additional work.

Operations delivers it.

Nobody tells finance.

Or the variation gets recorded but not approved.

Or approved but never invoiced.

Or invoiced months later and disputed because nobody remembers what happened.

You have done the work.

Consumed the capacity.

Paid the labour.

Purchased the materials.

And failed to create the corresponding revenue.

Create a simple variation process.

Identify.

Cost.

Approve.

Deliver.

Invoice.

Do not make it so bureaucratic that nobody uses it.

But do not rely on someone's memory three months later.

Problem 14: Management information arrives too late

If you discover that April's work was unprofitable while reviewing management accounts in June, you have a history lesson.

Useful, perhaps.

But late.

The earlier you can see margin deterioration, the more options you retain.

Perhaps managers need weekly information on:

Jobs exceeding labour allowance.

Projects below expected margin.

Unapproved variations.

Rework.

Overdue milestones.

Unexpected purchasing.

Overtime.

Major write-offs.

That does not mean producing full management accounts every Friday.

It means identifying the few leading signals that tell you profit is leaking before the leak becomes permanent.

ONS research into management practices found that firms with stronger management-practice scores were also more likely to use structured analysis in decision-making, and higher management-practice scores were positively associated with productivity. The survey covers UK firms with 10 or more employees and the ONS is careful to describe statistical association rather than proving that better management practices alone cause higher productivity.

Still, the management principle is useful:

Measure early enough to manage.

Problem 15: You do not know which jobs actually made money

This is surprisingly common.

Annual accounts say:

Profit: £250,000.

Fine.

Which jobs created it?

Which lost money?

Which customers?

Which service?

Which team?

Nobody knows.

Company-level profit can hide a lot.

Perhaps:

One division made £600,000.

Another lost £350,000.

Net result:

£250,000.

Without proper analysis, leadership concludes:

“We're profitable.”

Technically correct.

Strategically useless.

You need enough granularity to make decisions.

That might mean profitability by:

Job.

Customer.

Service.

Product.

Branch.

Team.

Channel.

Not every business needs every category.

Choose what helps explain the economics.

Do not chase another million pounds of turnover until you understand the last million

This is where sales growth can become avoidance.

Profit is disappointing.

The solution?

Sell more.

Maybe.

But imagine every £1 of additional sales currently generates only 3p of meaningful profit.

You could double revenue and create an enormous amount of operational pressure for a disappointing financial result.

Worse, if some work is genuinely loss-making, volume amplifies the loss.

There is a wonderfully simple principle here:

Do not scale a leak.

Find it first.

A simple margin-leakage audit

Take your last 20 to 30 completed jobs, projects, orders or customer engagements.

Do not only select the disasters.

Take a representative sample.

For each one, compare:

What did we sell it for?

Actual revenue after discounts, credits and adjustments.

What did we think it would cost?

Original labour.

Materials.

Subcontractors.

Delivery.

Other direct costs.

What did it actually cost?

Not assumptions.

Reality.

What changed?

Identify:

Extra labour.

Material variance.

Rework.

Scope changes.

Discounts.

Uncharged extras.

Overtime.

Special delivery.

Customer credits.

What gross profit did we expect?

Then:

What gross profit did we actually achieve?

Finally ask:

Why is there a difference?

Do this repeatedly and patterns usually emerge.

Perhaps estimating is consistently 15 per cent light on labour.

Perhaps one salesperson discounts heavily.

Perhaps one team creates most rework.

Perhaps one service is badly priced.

Perhaps one customer type absorbs enormous administration.

Now you have something actionable.

Do not search only for large problems

Profit leakage is often death by 100 cuts.

One per cent here.

Two per cent there.

Slight overtime.

Small discount.

Another site visit.

Unused materials.

Free delivery.

Minor credit.

A couple of missed billable hours.

Individually, none looks important enough for a meeting.

Collectively, they can turn:

15 per cent profit

into:

5 per cent profit.

That is why operations requires systems.

Not because every action needs controlling.

Because repeated small behaviour creates large financial outcomes.

Build a margin bridge every month

You should be able to explain why margin moved.

Suppose gross margin drops from 34 per cent to 29 per cent.

Why?

Price?

Sales mix?

Labour?

Materials?

Rework?

Discounting?

Overtime?

Customer credits?

You don't necessarily need an elaborate financial model.

You need management to be able to say:

“Roughly two points came from material costs, one point from overtime and two points from the larger proportion of lower-margin installation work.”

Now the number has meaning.

“Margin is down five points” merely describes the problem.

Metrics worth watching

Depending on the business, useful operational-profitability measures might include:

• Quoted gross margin
• Actual gross margin
• Contribution margin
• Estimate versus actual labour hours
• Material variance
• Overtime
• Rework cost
• Customer credits
• Unbilled variations
• Write-offs
• Productive utilisation
• Revenue per productive employee
• Gross profit per productive employee
• Average discount
• Profitability by customer
• Profitability by service
• Overhead as a percentage of gross profit
• Quote-to-job handover issues

Do not track all of them because I listed them.

Track the measures that explain where your money goes.

Be careful with generic margin benchmarks

Business owners love asking:

“What's a good margin?”

There isn't one universal answer.

A software company.

A construction contractor.

A recruitment business.

A retailer.

A manufacturer.

A consultancy.

All have completely different cost structures.

ONS publishes firm-level profit-margin and markup data from the Annual Business Survey and the figures show substantial variation between businesses and sectors. The latest dataset was released in December 2025 and covers data through 2024; the ONS classifies these as official statistics in development.

External benchmarks can provide context.

Your more important benchmark is:

What margin does our model need?

What margin have we historically achieved?

What changed?

What return is acceptable for the capital, risk and effort involved?

What not to do when profit is disappointing

Do not immediately cut every cost

Across-the-board cuts are easy.

Useful cost decisions are selective.

Removing the estimator who protects your margin to save salary might prove expensive.

Stopping management training because it doesn't directly generate revenue can preserve the inefficiency creating the problem.

Cut waste.

Not capability indiscriminately.

Do not automatically raise every price

Higher prices may absolutely be appropriate.

But if rework is the real problem, price rises merely ask customers to fund your inefficiency.

Fix both where necessary.

Do not blame employees before checking the system

If labour always overruns, ask why estimates are wrong.

If jobs always start late, inspect scheduling.

If materials are repeatedly wasted, inspect process and training.

Individual accountability matters.

So does management responsibility for the environment people operate inside.

Do not keep selling unsuitable work because the order book looks impressive

Turnover is not a trophy.

Do not wait for year-end accounts

By then, most operational decisions are ancient history.

A 30-day profit-recovery exercise

If the company is genuinely busy but profit is disappointing, I would spend the next month doing this.

Week 1: Establish where margin should be

Identify:

Current turnover.

Gross profit.

Contribution where relevant.

Overheads.

Operating profit.

Break-even.

Current sales mix.

Then establish whether the issue sits mainly in:

Gross margin.

Overhead.

Or both.

Week 2: Review completed work

Audit a representative sample.

Compare quoted versus actual.

Look for repeated variance.

Do not accept:

“That job was just a nightmare.”

Why was it a nightmare?

Week 3: Fix the biggest leaks

Perhaps:

Update pricing.

Change discount authority.

Improve scope control.

Introduce variation approval.

Update labour allowances.

Change scheduling.

Tighten purchasing.

Address repeated rework.

Reprice unprofitable customers.

Pick the few things most likely to matter.

Week 4: Build the early-warning system

Decide what managers need to see before month-end.

Assign ownership.

Create thresholds.

Review weekly.

Then keep reviewing until the numbers move.

Operations should understand margin

Profitability should not belong exclusively to finance.

Your project managers should understand how their decisions affect margin.

Sales should understand contribution.

Operations should understand the financial consequences of overtime, rework and poor scheduling.

Managers should understand the commercial cost of avoidable inefficiency.

That does not mean sharing every sensitive financial number with everybody.

It means giving people enough commercial understanding to make good decisions.

People cannot protect a margin they do not know exists.

Finance should understand operations too

The reverse is equally important.

A finance report can show labour overspend.

Operations may know:

The client changed access arrangements.

Equipment failed.

A supplier delay created idle time.

The original estimate was impossible.

Numbers identify where to look.

They do not replace operational understanding.

The strongest businesses connect both.

Finance knows what happened financially.

Operations knows what happened physically.

Together, you find why.

Better management practices matter

The relationship between management quality and business performance is not merely a coaching slogan.

ONS analysis from the Management and Expectations Survey found statistically significant relationships between stronger structured management practices and higher labour productivity among UK firms with 10 or more employees. The survey measures practices including monitoring, targets, incentives and the use of information in decision-making, although the data demonstrate association rather than proving a simple one-way causal relationship.

The useful takeaway for an owner is not:

Install more management bureaucracy.

It is:

Create enough visibility and discipline to understand what your operation is producing.

Systems beat repeated rescue

If every unprofitable project triggers:

A frantic meeting.

A bollocking.

A new spreadsheet.

A week of intense attention.

Then everything gradually returns to normal until the next bad job...

you haven't fixed anything.

Build the lesson into the system.

If scope caused the loss:

Improve scope control.

If quoting caused it:

Improve estimating.

If handover caused it:

Improve handover.

If scheduling caused it:

Improve scheduling.

If rework caused it:

Improve quality control and training.

If one customer caused it:

Review the customer economics.

This is agency.

Not:

How can we work harder next month?

But:

What keeps creating this outcome?

Winning work is only the first half of the commercial system

Sales matter.

Without customers, there is no business.

But winning the work only proves someone is willing to buy what you sell.

It does not prove you can deliver it profitably.

A commercially healthy business needs both sides:

Win good work.

Then:

Deliver good work with enough margin left behind.

When the order book is full and the profit is disappointing, resist the temptation to celebrate the first and vaguely complain about the second.

Trace the margin.

Quote to completion.

Estimate to actual.

Sales to delivery.

Customer to customer.

Find where the money is being lost.

Then fix the system producing the loss.

Because being busy is not the objective.

Turnover is not the objective.

Even growth is not automatically the objective.

The objective is building a business where the work you win produces enough value to justify doing it.