Yes, sometimes you should fire a customer.

But “they're a pain in the arse” is not enough information to make the decision.

The question is whether the customer still makes commercial sense once you account for:

• The revenue they generate
• The gross profit they contribute
• What it actually costs to serve them
• How reliably they pay
• How much management time they consume
• The operational disruption they create
• The risk they introduce
• The capacity they prevent you using elsewhere
• Their likely future value

A demanding customer can still be extremely valuable.

A friendly customer can quietly lose you money every month.

And a £500,000 account can be worse for the business than a £100,000 one.

So before terminating the relationship, diagnose it properly.

Do not fire a customer because they are difficult.

Fire them when the economics, behaviour or risk no longer make sense and there is no sensible way to repair the relationship.

Revenue is not customer profitability

This is where the conversation usually goes wrong.

Ask an owner who their best customers are and you will often hear the names of the biggest.

They spend the most.

Therefore they must be the most valuable.

Not necessarily.

Suppose Customer A spends £200,000 with you each year.

Customer B spends £100,000.

At first glance, Customer A appears twice as valuable.

But Customer A:

• Negotiates a 15 per cent discount
• Requires bespoke delivery
• Changes orders constantly
• Generates repeated rework
• Expects senior management involvement
• Pays 30 days late
• Raises frequent minor disputes
• Requires separate reporting
• Uses significantly more customer-service time

Customer B:

• Pays standard pricing
• Orders predictably
• Rarely causes rework
• Pays on time
• Uses your normal process
• Requires little management attention

Which is the better customer?

You cannot answer from revenue alone.

Research into customer profitability analysis and cost-to-serve has made this point for years. A 2008 case study and literature review in The International Journal of Logistics Management found that incorporating customer-specific service costs created a more complete view of customer profitability than looking primarily at product margins. The researchers were studying one Brazilian food-industry business, so the specific findings should not be generalised mechanically across every sector. The underlying accounting principle, however, is important: customers generating similar sales can create very different service costs.

Before firing anyone, work out what the relationship is actually worth.

Calculate the real cost to serve the customer

Start with revenue.

Then subtract the obvious direct costs.

That might give you gross profit or contribution.

But keep going.

Look at customer-specific costs such as:

• Discounts
• Rebates
• Delivery
• Returns
• Rework
• Warranty claims
• Bespoke reporting
• Additional administration
• Additional account management
• Special packaging
• Storage
• Rush orders
• Small-order processing
• Repeated site visits
• Extra quality checks
• Dedicated software or systems
• Credit-control time
• Financing the customer's payment terms
• Management escalation

Not every business will be able to attribute every pound perfectly.

You don't need an accounting exercise so complicated that nobody ever finishes it.

You need enough information to understand whether a supposedly important customer is actually making an acceptable contribution.

A simple example

Imagine a customer generates:

£250,000 annual revenue

Gross margin:

£75,000

Looks excellent.

But then you identify:

Additional discounts: £12,000

Bespoke logistics: £8,000

Rework and credits: £9,000

Dedicated administration: £6,000

Extra account-management resource: £10,000

Now the contribution looks more like:

£30,000

And we still haven't considered the amount of senior-management attention or capacity the account consumes.

That doesn't automatically make the customer bad.

£30,000 may still be worthwhile.

But it is a very different conversation from:

“We can't possibly lose them. They're worth a quarter of a million pounds.”

No.

They spend a quarter of a million pounds.

That is not the same thing.

Difficult and unprofitable are not the same problem

Separate them.

Difficult but profitable

They challenge you.

They negotiate.

They expect excellent service.

They notice mistakes.

They hold you accountable.

They sometimes make your team work harder.

But they pay appropriately for what they receive and the relationship creates healthy profit.

That might simply be a demanding customer.

Do not confuse high standards with being unreasonable.

Easy but unprofitable

Lovely people.

Never complain.

Everyone enjoys dealing with them.

Unfortunately, historic pricing means the work barely covers its direct cost.

That is a commercial problem even though nobody dislikes the customer.

Difficult and unprofitable

Now you probably have something worth addressing quickly.

Difficult, profitable and strategically important

That requires considerably more thought.

Especially if the customer creates access to an important market, meaningful future revenue, useful recurring work or strategic credibility.

Customer decisions should be commercial.

Not emotional.

First ask whether the customer is genuinely difficult

This question may sting slightly.

Are they difficult?

Or are you repeatedly letting them down?

Perhaps they chase because your company misses deadlines.

Perhaps they complain because quality varies.

Perhaps they ask for updates because nobody proactively communicates.

Perhaps every invoice is queried because the invoices are regularly wrong.

Perhaps their project manager keeps escalating because yours does not answer.

A customer repeatedly demanding that you honour what you promised is not necessarily difficult.

They may simply be unfortunate enough to have bought from you.

Before firing them, separate unreasonable demands from legitimate frustration.

Review:

What was promised?

What actually happened?

How often have you failed?

What complaints were justified?

What have they been forced to chase?

Would another reasonable customer react similarly?

Sometimes the best way to improve a difficult customer is to improve the service.

Then ask whether you created the commercial problem

Another uncomfortable possibility.

You agreed to the price.

You accepted the payment terms.

You allowed unlimited revisions.

You never defined the scope.

You offered free delivery.

You said yes to every special request.

You trained the customer that anything urgent would be prioritised.

Then eventually you became annoyed that they kept asking.

Businesses accidentally create difficult customers all the time.

Every exception teaches the customer how the relationship works.

If you answer the phone at 9pm repeatedly, availability becomes normal.

If every request for a discount works, negotiation becomes normal.

If scope changes are never charged, additional work becomes normal.

If payment at 75 days carries no consequence, late payment becomes normal.

The customer may simply be operating within a system you built.

Fix the system before blaming the person using it.

The first option is usually not firing them

Before ending the relationship, see whether you can change its economics.

That could mean:

Reprice the work

If the customer is profitable only because you ignore half the cost of serving them, price accordingly.

Maybe the relationship becomes perfectly acceptable at another price.

Tighten the scope

Define what is included.

Charge for additional work.

Introduce change controls.

Stop allowing everything to become “just a quick one”.

Change payment terms

Deposits.

Stage payments.

Shorter terms.

Direct debit.

Payment before delivery.

Whatever is commercially and contractually appropriate.

Change the service level

Perhaps the customer currently receives a premium service while paying a standard price.

Create clearer tiers.

Introduce minimum orders

If dozens of tiny orders create disproportionate processing cost, establish a minimum value or handling charge.

Reduce bespoke work

Move them towards your standard process where possible.

Change account management

Sometimes the commercial relationship is good but the personalities involved are not.

A different contact can transform it.

Set communication boundaries

Define escalation routes.

Response times.

Meeting frequency.

Decision-making responsibilities.

What genuinely qualifies as urgent.

The objective is to repair the relationship before discarding its value.

When repricing solves the problem

Sometimes a difficult customer becomes remarkably easy to tolerate when they are paying properly for the difficulty.

That sounds flippant.

It isn't.

Imagine a customer requires twice the project-management time of everybody else.

There are only three possibilities:

1. Reduce the project-management demand.
2. Absorb the additional cost.
3. Charge appropriately for it.

Businesses often choose number two accidentally.

Then resentment builds.

Pricing is partly about compensating the company for the resources a customer consumes.

If a customer requires a genuinely more expensive service model, there is nothing inherently wrong with charging for that model where your contractual and legal position allows it.

The problem is giving everyone a bespoke premium service while pretending you operate a standard one.

Late payment can change the relationship completely

This is particularly important for smaller businesses.

A £100,000 customer who pays reliably may be considerably more valuable than a £150,000 customer who requires constant chasing.

Because your business has already spent the money required to serve them.

Wages.

Materials.

Subcontractors.

Transport.

Overheads.

Then the customer holds your cash for another month.

The Office of the Small Business Commissioner advises businesses to agree payment terms clearly before work begins and to review customer payment processes, invoice promptly and act once payment becomes overdue. Its current guidance also suggests options including deposits and staged payments for larger projects where appropriate.

For UK business-to-business transactions, statutory late-payment rules may allow interest and fixed compensation where payment is overdue, depending on the contractual arrangements. GOV.UK states that statutory interest on qualifying late commercial payments is 8 percentage points above the Bank of England base rate where a different contractual interest rate does not apply.

That doesn't mean sending every customer an interest invoice the moment they reach day 31.

Commercial relationships require judgement.

But repeated late payment has a real cost.

Treat it accordingly.

Do not confuse contractual payment terms with late payment

There is an important difference.

Suppose you agreed 60-day terms.

The customer pays on day 59.

You may dislike the arrangement.

But they are not late.

You agreed to finance that payment period when you accepted the contract.

If 60 days is no longer commercially acceptable, renegotiate the terms for future work.

The Small Business Commissioner's contract guidance strongly encourages businesses to understand and negotiate payment terms before supplying work and to record what has been agreed.

Do not agree to terrible terms because you desperately want the customer and then act surprised when they use them.

When should you seriously consider firing the customer?

There is no universal threshold, but several situations should trigger a proper review.

1. The relationship is structurally unprofitable

Not one bad job.

Not one mistake.

The relationship itself repeatedly loses money.

You have repriced where possible.

Changed the service model.

Controlled scope.

Improved your own efficiency.

And the economics still don't work.

At some point, revenue that destroys profit is not helpful revenue.

Keeping it because losing turnover feels frightening can make the overall business weaker.

2. They consistently fail to pay

Occasional administrative mistakes happen.

Persistent late payment is different.

Especially where customers:

Ignore reminders.

Create spurious disputes after invoices become due.

Continually move payment dates.

Change requirements for processing invoices after work is complete.

Make repeated promises they do not honour.

Expect you to continue supplying despite significant arrears.

The Small Business Commissioner identifies excessive delays in resolving invoice disputes and repeatedly poor payment behaviour among the practices that can create serious problems for suppliers.

Before terminating the relationship, follow appropriate contractual and debt-recovery processes.

But do not allow a customer to turn your balance sheet into their overdraft indefinitely.

3. They abuse your employees

This one changes the calculation.

Customers can be frustrated.

They can complain.

They can challenge.

They can expect errors to be fixed.

They cannot reasonably expect to abuse your staff.

Threats.

Personal insults.

Harassment.

Discriminatory abuse.

Repeated intimidation.

Aggressive behaviour.

No amount of gross margin automatically justifies subjecting employees to that.

Establish what happened.

Respond proportionately.

Give clear boundaries where appropriate.

But leadership sends a very powerful message when employees repeatedly experience abuse and are told:

“Unfortunately they're a big account.”

You are teaching your people precisely where they sit in the hierarchy.

4. They pressure you to do something unsafe, unlawful or unethical

Walk away.

A customer wanting you to:

Ignore safety requirements.

Falsify documentation.

Misrepresent something.

Circumvent regulation.

Hide defects.

Backdate records.

Cut a legally required corner.

is not a valuable customer.

They are a liability with a purchase order.

Commercial pressure does not transfer responsibility away from your business.

5. Scope creep has become the business model

A small extra request occasionally is normal.

Continuous uncharged extra work is not.

If the customer expects 120 per cent of the service for 100 per cent of the price and refuses every attempt to clarify scope, you have a structural problem.

Either change the terms or reconsider whether you want the work.

6. They consume scarce capacity that could generate substantially better returns elsewhere

This is the opportunity-cost problem.

Suppose a customer produces £40,000 contribution per year.

Fine.

But servicing them absorbs a specialist team operating at full capacity.

You are currently rejecting or delaying customers that could create £100,000 of contribution using the same resource.

The first customer is technically profitable.

They may still be economically expensive.

This becomes particularly relevant when a business reaches capacity.

When spare capacity exists, marginally profitable work may be worth taking.

When capacity becomes scarce, allocation matters considerably more.

7. Their risk has become unacceptable

Perhaps their financial health is deteriorating.

Perhaps outstanding balances keep growing.

Perhaps contractual exposure has increased.

Perhaps the service itself has moved into an area where your company cannot confidently deliver.

Perhaps the customer creates reputational or regulatory risk.

Risk belongs inside the profitability decision.

An account can look attractive right until the downside materialises.

8. The relationship no longer fits the company you are building

Businesses evolve.

Perhaps you are moving away from tiny bespoke projects towards recurring contracts.

Perhaps you are withdrawing from a sector.

Perhaps one historic service no longer fits the business.

Perhaps the customer's requirements depend on processes you deliberately want to retire.

Not every customer relationship needs to exist forever.

Sometimes neither party has done anything wrong.

The fit simply disappeared.

But beware of firing customers while desperate for capacity

There is a fashionable piece of business advice that goes something like:

Fire the bottom 20 per cent of your customers.

Why?

Because somebody once put it in a slide deck?

Customer profitability is not that neat.

A lower-margin customer may:

Fill otherwise unused capacity.

Provide predictable recurring revenue.

Refer highly valuable customers.

Buy additional products later.

Require very little acquisition cost.

Have strong future potential.

Balance seasonality.

Carry strategic importance.

Do not mechanically rank customers in Excel and terminate whichever names appear at the bottom.

Understand why they are there.

Then decide.

Customer lifetime value matters too

A newly acquired customer can temporarily look less profitable because onboarding costs are front-loaded.

Equally, an old customer may currently generate high margins only because nobody has noticed how much additional support they now require.

Profitability needs a sensible time horizon.

Academic work on customer profitability has long distinguished between current-period profitability and the wider economic value of the customer relationship.

Do not fire somebody after one awkward quarter if the long-term economics remain attractive and the underlying issue is fixable.

But do not keep a structurally dreadful account for ten years because:

“They've always been a good customer.”

History doesn't pay next month's payroll.

Beware customer concentration before making the decision

This is where the intellectually correct decision can become commercially reckless.

Imagine the difficult customer represents 35 per cent of your turnover.

You calculate the account properly.

Margins are mediocre.

Management hates servicing it.

You conclude the business would be better without them.

Maybe it would.

Eventually.

But removing 35 per cent of revenue next month without replacement demand or sufficient reserves could damage the business more than the customer currently does.

This is why customer concentration creates dependency.

Sometimes the correct decision is not:

Fire them tomorrow.

It is:

Spend the next twelve months reducing our dependence so that we have a genuine choice.

Build other accounts.

Strengthen the pipeline.

Improve cash reserves.

Reallocate capacity gradually.

Renegotiate the existing relationship.

Then make the decision from strength rather than frustration.

Agency requires options.

Your sales team may hate this conversation

Understandably.

Salespeople are usually measured on sales.

So a £500,000 customer looks brilliant.

Operations may see the same customer completely differently.

Finance sees late payments.

Operations sees repeated changes.

Customer service sees complaints.

The MD sees escalation.

Sales sees:

£500,000.

Neither perspective is necessarily dishonest.

They are measuring different things.

This is exactly why customer profitability should not live only inside sales reporting.

A healthy commercial review should consider:

Revenue.

Margin.

Cost-to-serve.

Payment performance.

Capacity.

Risk.

Retention potential.

Growth potential.

You need the whole relationship.

Create a customer health scorecard

You do not need to build a ridiculous 42-column spreadsheet.

For significant accounts, regularly review perhaps six things.

Revenue

What do they buy?

Profitability

What do we actually make?

Cost-to-serve

What unusual resources do they consume?

Payment performance

Do they pay as agreed?

Relationship health

Is the relationship constructive and sustainable?

Strategic value and risk

What does this account mean beyond this month's invoice?

That simple review can expose customers that look fantastic in a turnover report but less impressive elsewhere.

Before firing them, run the repair conversation

If the relationship still has potential, talk.

Not emotionally.

Not:

“Your company is a nightmare to deal with.”

Explain the commercial issue.

For example:

The service requirement has expanded materially beyond the original scope.

The current pricing no longer reflects the resources required.

Payment performance is creating an unsustainable exposure.

The current turnaround expectation cannot be maintained reliably.

The volume of low-value orders is creating disproportionate administration.

Then propose a solution.

Different pricing.

Different scope.

Different terms.

Different ordering process.

Different service level.

You may discover the customer is perfectly willing to change.

Remember, they may have no idea they are unprofitable for you.

They only know what you have been willing to provide.

What if they say no?

Then you have considerably better information.

You tried to repair the economics.

They rejected the alternatives.

Now the choice becomes clearer.

You can consciously accept the relationship as it stands.

Or end it.

But you are no longer quietly resenting an arrangement neither party has ever attempted to change.

How do you fire a customer professionally?

First, decide whether you are:

Stopping immediately.

Giving notice under an existing agreement.

Not renewing a contract.

Finishing current commitments but declining future work.

Changing the commercial terms and allowing the customer to decide whether to continue.

Those are very different situations.

Then check the contract.

The Small Business Commissioner's contract guidance notes that contractual breaches can vary in seriousness, with some potentially allowing the other party to terminate or claim compensation. For significant relationships or unclear termination rights, proper legal advice is sensible.

Do not assume that because you no longer like the customer you can simply abandon an existing contractual obligation.

Check the legal position before terminating services

This article is about commercial management, not legal advice.

Particularly in consumer-facing businesses, additional legal considerations can apply.

The Consumer Rights Act governs consumer contracts, including requirements around fair and transparent contractual terms. The Competition and Markets Authority updated its unfair-contract-terms guidance in July 2026.

Service providers must also comply with equality law. Section 29 of the Equality Act 2010 prohibits service providers from discriminating in the provision or termination of services, and the updated statutory Code of Practice for services came into force on 5 August 2026.

There may also be sector-specific regulation, contractual duties or competition-law considerations. GOV.UK's competition-law guidance notes, for example, that businesses in a dominant market position can create competition-law risk by refusing to supply an existing customer without objective justification.

For an ordinary SME ending a normal B2B commercial relationship, many of those issues will not apply.

But the principle is simple:

Check before acting where the legal position is material.

Especially if the relationship is large, regulated, contracted or contentious.

Do not make the exit personal

Your customer does not need a character assessment.

You do not need to explain that half the team hates dealing with them.

Keep it commercial.

Perhaps:

The service is no longer something your business will be providing.

The existing arrangement is no longer commercially sustainable.

Your capacity is being redirected.

The current requirements no longer fit your delivery model.

You are unable to continue under the existing commercial terms.

Give whatever notice is contractually and professionally appropriate.

Complete outstanding obligations properly.

Deal with outstanding invoices.

Arrange handover where required.

Remain professional.

Today's difficult customer may become tomorrow's referrer, supplier, acquirer or completely reasonable customer under another set of circumstances.

You gain nothing by setting fire to the bridge on your way across it.

Do not celebrate losing revenue

There is another weird cultural habit in business.

Someone finally fires an awful customer.

Everyone cheers.

Good.

Maybe it was absolutely the right decision.

But losing meaningful revenue still creates work.

What capacity has been released?

What contribution has disappeared?

What happens to employees previously servicing the account?

What does the pipeline look like?

How quickly can replacement work be won?

What was the customer concentration before and after?

A commercially correct decision still requires a plan.

The objective is not fewer customers.

It is a stronger customer portfolio.

Look at your customer portfolio, not just individual accounts

One problematic customer matters.

But the wider question is:

What type of customer does your business keep attracting?

If you repeatedly acquire:

Price-sensitive customers.

Late-paying customers.

Tiny customers requiring enormous support.

Projects outside your core capability.

Buyers who demand excessive customisation.

Then firing one customer will not solve much.

Marketing and sales may be feeding the company work it should never have accepted.

That connects customer management directly back to:

Target market.

Marketing.

Qualification.

Sales.

Pricing.

Contracting.

Capacity planning.

The customer you eventually fire may simply be the final symptom of a decision made months earlier in the sales process.

Sales qualification should protect operations

One of the most powerful things a business can do is define the characteristics of a bad-fit customer before they become one.

For example:

Minimum order values.

Acceptable payment terms.

Geographic limits.

Project types.

Technical requirements.

Minimum margin.

Capacity requirements.

Credit risk.

Service expectations.

Then allow salespeople to qualify accordingly.

Otherwise the business celebrates every sale and audits the damage afterwards.

Not every customer should become your customer.

That is not arrogance.

It is commercial discipline.

A practical customer-exit test

Before ending a significant relationship, answer these questions.

1. Is the customer actually profitable?

Calculate it properly.

2. Is the problem fixable?

Pricing?

Scope?

Terms?

Process?

Account management?

3. Have we caused part of the problem ourselves?

Be honest.

4. What future value could we lose?

Recurring revenue?

Cross-sales?

Strategic value?

Referrals?

5. What capacity would we release?

And what could realistically replace the work?

6. How dependent are we on this customer?

What happens to cash and profit if they disappear tomorrow?

7. What contractual or legal obligations exist?

Check them.

8. Have we attempted a sensible commercial reset?

Unless circumstances make immediate termination necessary.

9. What happens if nothing changes?

Run that forward 12 months.

How much money, time and capacity does the relationship consume?

10. Would we knowingly take this customer on today?

This is probably my favourite question.

Imagine they were not already a customer.

Knowing everything you now know about:

Their margin.

Their behaviour.

Their requirements.

Their payment.

Their risk.

Would you deliberately sign them?

If the answer is:

Absolutely not.

You need a very good reason to continue simply because you already did.

Sometimes firing the customer is really firing your old business model

This is worth thinking about.

A customer can remain from an earlier version of the company.

Back when you:

Took any work.

Priced differently.

Had spare capacity.

Needed credibility.

Were entering the market.

Did everything bespoke.

Personally handled every customer.

That customer may have helped the business enormously at the time.

You can be grateful for that.

And still recognise that the relationship no longer works.

Businesses change.

Customers change.

Commercial relationships can finish without anybody being the villain.

Sometimes you should keep the difficult customer

Equally, there are customers I would not rush to lose.

The one who exposes weaknesses in your processes.

The one who pushes standards higher.

The one who provides incredibly useful feedback.

The one whose difficult requirements are properly reflected in margin.

The one temporarily going through a problem after years of excellent behaviour.

The strategic customer whose current economics make sense over a longer horizon.

The one your company has genuinely let down.

Do not build a business full of customers who never challenge you.

Comfort is not the same as quality.

The objective is not easy customers.

It is good commercial relationships.

The customer is not always right

But neither is the business.

A healthy commercial relationship needs value on both sides.

The customer needs to believe what they receive is worth what they pay.

You need to believe what they pay justifies what it takes to serve them.

When that balance disappears, do not simply tolerate it until everybody resents one another.

Look at the numbers.

Look at the behaviour.

Look at the risk.

Change the relationship if you can.

And if you cannot, be prepared to end it professionally.

Revenue is useful.

Profit is better.

But a genuinely good customer creates something more valuable still:

A sustainable exchange that both sides remain happy to continue.