Here is a number that decides more funding conversations than founders expect: what percentage of your revenue comes from your single biggest customer? It gets asked early, it gets asked in almost every diligence process, and the answer reshapes the valuation more than most founders realise. If you cannot say the number out loud, that is itself the finding.

What concentration risk actually is

Customer concentration risk is the exposure created when too much of your income depends on too few customers. A startup billing £240,000 a year with one client providing £150,000 of it is not a £240,000 business in risk terms. It is a £90,000 business with a £150,000 option attached, and somebody else holds the option. Losing that client removes 62.5% of revenue on whatever notice the contract specifies — often thirty days, sometimes none at all.

The distinction matters because the two businesses look identical on a profit and loss account. Concentration never appears as a line in your accounts. It only appears when someone asks for revenue split by customer, which is exactly why the question comes so early in diligence.

Why investors, lenders and acquirers all check it

  • Investors are pricing the durability of the revenue, not its size. A diversified £240,000 of income is worth materially more than a concentrated £240,000, because the probability distribution of next year's revenue is completely different.
  • Lenders are testing whether you can still service the debt after your worst plausible month. One contract ending should not put a loan into default, and a lender models exactly that.
  • Acquirers discount hard, and often structure around it: an earn-out that pays only if the major client stays, or a chunk of the price held back for a retention period. Concentration does not just cut the headline number, it changes how much of it you actually receive.

It also costs you money every day before any of those conversations happen. A client providing most of your revenue knows it, and that knowledge sits inside every price negotiation, every payment-terms discussion and every unreasonable request you absorb because you cannot afford to lose them.

The thresholds worth watching

There is no statutory threshold and no regulator's line here — this is commercial judgement, and it varies by sector. As a working set of tripwires: a single customer above 25% of revenue belongs on the board agenda; above 40% it is the board agenda. A useful second measure is the top-three share, because three clients at 20% each is a genuinely different risk from one at 60%, and a single revenue-concentration figure hides that.

Know the number before you are asked Clean management accounts show revenue by customer at a glance. Being able to say "our largest client is 18% of revenue and falling, and our top three are 41%" signals a founder in control of the business. Being surprised by the question signals the opposite, and it is remembered. Our runway and forecasting guide covers building the reporting that makes this routine.

A worked example: what losing the big client really costs

Illustratively, take a two-year-old agency. Revenue £240,000. Fixed monthly costs of £14,000 covering two salaries, software and an office. Cash in the bank £38,000. The largest client, a retail group, is on £12,500 a month — £150,000 a year, or 62.5% of revenue — with thirty days' notice.

The retail group gives notice in March. Work through what happens:

  • Monthly revenue drops from £20,000 to £7,500 against £14,000 of fixed costs. The business goes from £6,000 a month positive to £6,500 a month negative, a £12,500 swing.
  • £38,000 of cash divided by a £6,500 monthly burn is 5.8 months of runway, and that is before the final invoice goes unpaid.
  • It does go unpaid. The last two months, £25,000 plus VAT of £5,000, is £30,000 outstanding from a client who has already left.

Three things are recoverable here, and founders routinely claim none of them:

  1. Statutory interest. Under the Late Payment of Commercial Debts (Interest) Act 1998 you can charge 8% above the Bank of England base rate on a late business-to-business invoice, unless your contract sets a different rate. The base rate has stood at 3.75% since the Monetary Policy Committee's hold on 30 July 2026, the fifth consecutive meeting at that level, so statutory interest runs at 11.75%. On £30,000 that is £3,525 a year, or £9.66 a day. Forty-five days late is £434.70.
  2. Fixed debt recovery compensation. On top of the interest, late payment legislation lets you charge a fixed sum per invoice: £40 on debts up to £999.99, £70 from £1,000 to £9,999.99, and £100 on debts of £10,000 or more.
  3. VAT bad debt relief. You have already paid HMRC the £5,000 of VAT on an invoice nobody paid. Once the debt is six months past its due date and you have written it off in your day-to-day VAT accounts and transferred it to a separate bad debt account, you reclaim the £5,000 on your VAT return. The claim window runs for four years and six months from the later of the due date and the date of supply, so this is recoverable long after most founders have written it off mentally.

None of that saves the business. It is £5,534.70 against a £150,000 hole. What it does is buy most of another month of runway for the cost of three emails, and the month you buy is the month you spend selling.

What to do about it, in the order that works

  1. Measure it monthly, not annually. Concentration creeps up silently whenever your best client grows faster than everyone else. Put revenue by customer and top-three share in the monthly management pack so the trend is visible before the level is dangerous.
  2. Fix the notice period before you fix the mix. Moving your largest contract from thirty days' notice to ninety does not reduce concentration by a single percentage point, but it triples the time you get to react. It is usually the cheapest and fastest thing on this list, and large clients rarely object because it suits their planning too.
  3. Stage-gate the growth. Set a rule at board level — no single client goes above 35% of forecast revenue without a matching plan to grow the base. The rule stops the comfortable version of this problem, which is the client who keeps buying more.
  4. Convert one-off work into recurring revenue across the smaller accounts. A broader base of retained clients does more for durability than one more project win.
  5. Get the payment terms right and enforce them. The law makes a payment late 30 days after the invoice or delivery where nothing is agreed, and caps agreed business-to-business terms at 60 days unless a longer period is genuinely fair to both sides. If your biggest client pays on 90 days, they are financing themselves with your cash and multiplying the damage of the day they leave.
  6. Check whether your major clients hold a Fair Payment Code award. The Small Business Commissioner's code, which replaced the Prompt Payment Code in December 2024 and now has over 450 signatories, awards Gold to businesses paying at least 95% of invoices within 30 days, Silver for 95% within 60 days including 95% of small-business invoices within 30, and Bronze for 95% within 60 days. It is a free, public signal about who will actually pay you on time.

The founder takeaway

Growth that comes entirely from one customer feels excellent and is quietly the most dangerous shape a young business can take. The startups that raise well and sell well are the ones that saw the concentration on a management report and diversified before the market forced them to. We build the monthly reporting that makes the number impossible to miss, and we make sure that when an invoice does go bad you claim the interest, the fixed sum and the VAT back rather than absorbing all three. Our corporation tax guide covers how bad debts land in the tax computation. Get started.