Client Concentration: What to Do When One Client Is Most of Your Income

Published July 30, 2026

Disclosure: Some links below are affiliate links. If you sign up through them we may earn a commission at no extra cost to you. We only recommend tools we'd genuinely suggest to a friend. See our full disclosure.

Freelance advice has one direction of travel, and it is upward. How to find clients, how to write the cold email, how to win the proposal, how to onboard well, and — the most profitable advice of the lot — how to turn one project into repeat work so you stop starting from zero every month.

All of it counts clients up. None of it asks what share of your income any single one holds.

That is a strange omission, because the advice actively pushes you toward the answer. Retention is genuinely cheaper than prospecting. Repeat clients genuinely are more profitable. Follow that entirely sound guidance for a year or two and you arrive somewhere nobody described: one client, most of the money, and a business that has quietly stopped being a business.

The quick version

• Concentration is not a decision anyone makes. It accumulates out of saying yes to the easiest available work, which is why it needs a scheduled check rather than a moment of judgement.
• The percentage is the wrong test. The real question is whether you could survive without that client for as long as it takes you to replace them.
• The cost arrives before the client leaves, in every place you stop enforcing scope, stop raising the rate, and stop saying no.
• Three shapes of concentration exist and they are not equally fragile — many projects, one long project, or one contact person inside a company. The third is the most dangerous and the least noticed.
• A notice period is the cheapest insurance available. It converts a cliff into a ramp, and asking for one protects the client too.
• “Get more clients” is right and too slow to be a first move. The fast moves are the contract clause and the rate.
• Deliberate concentration is fine. Accumulated concentration is what hurts, because nothing about it was ever chosen.

The number nobody computes

Take your last six months of income and work out what percentage came from your single largest client. It takes about five minutes with whatever you already use to track income, and most freelancers have never once done it.

The reason it goes uncomputed is not laziness. It is that income arrives one invoice at a time, and every individual invoice looks like a success. The composition of the total is invisible unless you deliberately go and look at it, and nothing in the ordinary run of a week prompts you to.

Do it and you will land in one of three places. Under a third from any one client, and you have a portfolio. Between a third and a half, and you have a business with a large customer, which is normal and worth watching. Over half, and the arithmetic has changed underneath you: you no longer have several clients, you have one client and some others.

Why the percentage is the wrong test anyway

It is tempting to reach for a threshold — a number above which you are exposed and below which you are fine. Every such number is invented. Treat any of them as a rule and you are taking a guess seriously because it arrived with a decimal point attached.

The useful test is one you can genuinely answer about your own situation, and it has two halves.

How long could you last? If that client stopped at the end of this month, how many months would the remaining income plus whatever you have saved actually cover?

How long would replacing them take? Not how long you would like it to take. How long it took the last time — from first contact to first payment, including the weeks a serious enquiry usually spends waiting on someone else’s decision.

If the second number is larger than the first, you have a concentration problem, whatever the percentage says. If the first is comfortably larger, you have a large customer and a manageable risk. This is a better test because it accounts for the two things a percentage cannot see: your savings, and the length of your sales cycle. Someone at seventy percent with six months of costs in the bank and a fast pipeline is genuinely safer than someone at forty percent with neither.

The tax you are already paying

Here is the part that makes this urgent rather than theoretical. Most people picture concentration risk as an event in the future — the day the email arrives. In practice the cost is already being charged, monthly, and it does not appear on any invoice.

Dependence changes behaviour before it changes income. Look for these, honestly:

• You enforce scope with everyone except them. The extra request that would get a polite “that’s outside what we agreed, here’s what it would cost” from any other client gets absorbed from this one.
• Their late payment gets a softer follow-up. You would chase anyone else on the day. Them you give a week, and then a gentle note, because the relationship feels too valuable to risk over timing. (Everything in how to invoice a client still applies; you are simply not applying it here.)
• You have not raised their rate in a long time. And when you consider it, there is a flicker of genuine fear. That flicker is the whole article in one sensation.
• You schedule around their availability. Other work waits on their calls, their feedback, their quiet fortnight in August.
• You have said yes to work outside what you actually sell, because a single no felt disproportionate to the relationship.
• You read their messages for tone. Wondering whether this one sounds shorter than usual is a specific and unmistakable symptom.

Every one of those is a discount. None of them is invoiced. Individually each is a small, reasonable accommodation to a good client — which is exactly why the total never gets added up. And notice that the site’s usual home for these behaviours is difficult clients and raising your prices, as though they were problems of assertiveness. Sometimes they are. Often they are a structural position expressing itself through your inbox.

Three shapes of concentration, in order of fragility

“One client is sixty percent of my income” describes at least three different situations with different failure modes.

One client, many separate projects. They come back repeatedly and choose you each time. This is the sturdiest version: no single project ending takes the income with it, and the repeated choosing is real evidence of value. It is still concentration, but it degrades gradually rather than stopping.

One client, one long engagement. A retainer, or a project that keeps extending. The income has an end date built into it whether or not anyone has said so, and the danger is that a long engagement feels more permanent the longer it runs, when the opposite is true — every month makes the eventual end nearer, not further away.

One client, one person inside it. This is the fragile one, and almost nobody counts it. Your relationship does not really exist with the company; it exists with the person who hired you, vouches for you internally and approves your invoices. If they change jobs, go on extended leave, or get reorganised into a different team, you are a line item to a stranger who did not choose you. The client red flags piece makes the case for finding out who signs off before you take a project on; the same question applies to the client you have had for two years, and the answer changes over time without anyone telling you.

It is worth naming which of the three you actually have, because the mitigations differ. The first needs a pipeline. The second needs a notice period. The third needs a second relationship inside the same company — which costs one introduction and is the single highest-return thing on this page.

Where the risk actually sits, which is not your work

The instinctive mitigation is to be so good they never leave. It is a comforting plan and mostly beside the point, because most of the reasons a large client stops have nothing to do with the quality of what you delivered.

Budgets are set annually and cut mid-year. The person who championed you moves on. A merger arrives with an existing supplier list. Their own market has a bad quarter and every discretionary line goes. A new head of department wants their own people, which is not a judgement on you and will feel like one. Or the project simply finishes, successfully, as it was always going to.

None of that is fixable with better work, and that is the point: excellence is not a hedge. It is what earns you the position. It does not protect you inside it.

What to do about it, in order of speed

The standard answer is “get more clients”. It is correct, and as a first move it is nearly useless, because it takes months and you are looking for something to do this week. So take these in order.

1. Get a notice period in writing. This is the fastest, cheapest and most overlooked move available. A thirty or sixty day notice clause turns a cliff into a ramp — instead of income stopping, you get paid for a period during which you can prospect properly. Ask for it at a natural moment: a renewal, a new statement of work, the start of the next project. It is easy to ask for because it is genuinely mutual — they also want notice from you, and framing it that way is honest rather than tactical. If you do not have a written agreement to add it to, how to write a freelance contract covers what else should be in there.

2. Raise their rate. Counterintuitive, and it is the second-fastest move for a reason. Concentration and underpricing arrive together, because the rate you are most afraid to touch is the one you have left alone longest — and the one you have left alone longest is the client you depend on most. Bringing them up to what you charge new people does two things at once: it increases the income you would need to replace, and it tells you something true about the relationship’s strength, at a time you chose rather than one they chose. If they leave over a fair, well-explained increase, the dependence was worse than you thought and you have found out on the best possible terms. How to raise your prices covers the wording; how to price freelance services covers where the number should be.

3. Build a second relationship inside the same client. If your position depends on one contact, one introduction changes the whole risk profile. It is a small, ordinary thing to ask for — a call with the person who will use the work, a copy of the report going to their manager, an intro to the neighbouring team with a similar problem. Nothing about it is political. It just means that when your champion leaves, someone else in the building knows what you do.

4. Make the second client a standing arrangement, not an emergency. The goal here is not a second large client — that takes as long as replacing the first. It is a small ongoing commitment that exists specifically so the alternative to your main client is not zero. A modest monthly retainer, a recurring small piece of work, anything with a pulse. It is worth more than its invoice value because of what it does to your willingness to say no. It also fixes something a large client cannot: money that arrives on a predictable schedule rather than in project-shaped lumps, which is the structural half of freelance cash flow.

5. Keep the smallest possible prospecting habit alive permanently. Not a campaign — a habit. One conversation a week, or an hour on a Friday. The quiet spell that follows a full calendar is almost always caused by switching marketing off during the full calendar, which is the same mechanism described in what to do when you have too much work, operating on a longer timescale.

6. Package what you already do into something that sells without you. You have, by definition, done one thing repeatedly and well. That is unusually good raw material for a template, a guide or a small digital product — income that does not depend on any single relationship. It is the slowest item here, which is why it is last, and it is the only one that changes the shape of the business rather than the safety of the current one.

If the paperwork side of this is what keeps getting postponed — the agreement with the notice clause in it, the rate-increase note, the follow-up on a late invoice — my Freelancer’s Client Toolkit has the proposal, agreement, onboarding, invoice and payment-follow-up templates already written, so adding a notice period at the next renewal is a five-minute edit rather than an evening spent drafting.

When concentration is the right call

None of this argues for spreading yourself thin on principle. Deliberate concentration is often the best available position, particularly early on.

One reliable client who pays on time is a far better place to be than five who do not, and the stability is worth real money — it is what makes it possible to plan, to stop taking bad-fit work, and to build something on the side that eventually replaces the dependence entirely. A well-paid anchor client funding the year it takes to build a product is a strategy, not a failure.

The difference is not the percentage. It is three features that a chosen concentration has and an accumulated one never does:

• You know what it is buying you. Time, stability, a portfolio, the runway to build something else.
• It has a date or a condition on it. “Until the product launches.” “Through this year.” “Until the second retainer starts.” Something that makes it a phase rather than a permanent state.
• The contract has a notice period. Because choosing to depend on someone is a reasonable decision, and choosing to do it without protection is a different decision that nobody makes on purpose.

Accumulated concentration has none of those, and the reason is worth being kind to yourself about: nobody ever decided it. You said yes to the good client, then yes again, then yes to the extension, and each individual yes was the right call. The position built itself out of correct decisions, which is precisely why it needs a scheduled review rather than better judgement in the moment. Put it in the same monthly slot where you already check where each project stands, and it becomes one line: what percentage, and is that still what I want.

The decision, in order

1. Compute the number. Six months of income, largest client, percentage. Five minutes.
2. Run the two-part test. Months of survival without them, against months to replace them.
3. Name the shape. Many projects, one long engagement, or one person — and mitigate the one you actually have.
4. Audit your own behaviour for the six symptoms. They are the cost you are already paying.
5. Add a notice period at the next natural moment.
6. Raise the rate if it has drifted below what you charge new clients.
7. Get a second contact inside that client, if the risk is one person.
8. Start one small standing arrangement elsewhere, so the alternative to them is not zero.
9. Diarise the check so it happens again in three months without you remembering to.

The bottom line

A freelance business with one dominant client is not really a business. It is a job, with none of a job’s protections — no notice, no redundancy terms, no employment rights — and it was arrived at by following advice that was correct at every individual step.

That last sentence is a metaphor here, and sometimes it stops being one. Concentration is a question about the share of your income; whether the working relationship has itself taken the shape of employment is a separate question about how the work is actually conducted — and you can have either without the other.

The failure mode is not the day they leave. It is the eighteen months before, during which the dependence has already been quietly setting your prices, choosing which work you accept, and deciding what you are willing to say. That cost is being charged now, whether or not the relationship ever ends.

You do not fix it by being less loyal to a good client, and you certainly do not fix it by finding fault with one who has done nothing wrong. You fix it by making the position deliberate: know the number, put a notice period behind it, charge what the work is worth, and keep the smallest possible door open somewhere else. Do that, and a large client becomes what they should have been all along — the best thing in your business, rather than the only thing in it.

Next: how to find freelance clients, how to raise your prices, and how to end a client project.