Skip to content
AdvancedMetric 31 of 35

Revenue Concentration Risk

How much of your MRR sits with a handful of accounts.

Abderrahim Ben Jazia, PhDBy · Founder of FlowMRR, previously built and sold AIS (Avance Immédiate Services), French tax-credit softwareUpdated:

Put this metric to work: explore FlowMRR's Stripe MRR analytics to track revenue movements, retention and forecasts from your Stripe account, or see how FlowMRR compares as a Baremetrics alternative for Stripe.

How evenly your MRR is spread across customers, with one value per customer (records sharing an email count once) and paying customers only. FlowMRR grades the risk from the largest customer's share and the top five's: critical from 25% or 60%, high from 15% or 45%, moderate from 8% or 30%, low below; it gives no verdict under 20 paying customers. The Gini coefficient (0 for equal revenue, close to 1 for extreme concentration) and deeper shares are shown but not graded. Two businesses at $50,000 MRR can have very different risk profiles: one where the top customer is 2% of revenue, another where it is 40%.

Formula

Largest-customer share = Largest customer's MRR ÷ Total MRR; Top-5 share = MRR of the 5 largest customers ÷ Total MRR

Worked example

$80,000 MRR, top customer pays $18,000/mo (22.5% of MRR), top 5 customers pay $38,000/mo (47.5%) → high concentration: the largest share is above 15% and the top five above 45%, both under the critical marks. Losing that one account instantly drops MRR by nearly a quarter.

Other definitions you will see

Other reports measure concentration with the top-10 or top-20 share, the Herfindahl-Hirschman index (the sum of squared shares) or the Gini alone, on annual revenue or contract value instead of current MRR, or per Stripe customer record, which reads lower when one buyer pays through several records.

Illustrative comparisons

Top-5 share of MRR

These unsourced ranges are illustrative scenarios, not measured industry benchmarks, percentiles or recommended targets. Their qualitative labels are editorial interpretations. Compare your own cohorts using the same segment, period, currency and measurement basis.

Diversified
< 30%
Watch
30–45%
High risk
> 45%

Why it matters

Concentration is invisible in a headline MRR number — a business can be growing nicely while quietly becoming a two-customer company. Investors run this exact stress test in diligence ("what happens to ARR if your top account leaves?"), and it should change how you negotiate contracts, set payment terms, and prioritise support for those accounts.

Common mistakes

  • Only looking at logo count, not revenue share — 5% of customers can easily be 40% of MRR.
  • Never re-running the check after a big new enterprise deal, which is exactly when concentration spikes.
  • Treating concentration as purely bad — a few large, sticky accounts can also mean real enterprise traction. The point is to know the number, not to always minimise it.

Calculating it from Stripe data

Concentration is measured on the current MRR snapshot, with one value per customer. Start from headline MRR per subscription (annual plans ÷ 12, forever discounts deducted, trialing, incomplete, incomplete_expired and paused excluded), sum it per customer and keep only customers with positive MRR. Grouping is the step that changes the answer: one organisation can pay through several Stripe customer records, one per team, legal entity or workspace. FlowMRR groups records by email, so records with different emails that belong to the same company stay separate; if you sell to groups, consolidate those accounts before reading the top-customer share.

Each share is a customer's MRR divided by the account's total MRR. From the sorted list you get the largest customer's share, the top five's share, the number of customers that make up half of MRR, the Herfindahl-Hirschman index (the sum of squared shares, which weights large customers heavily) and the Gini coefficient. FlowMRR reports all of them but grades risk from the top-1 and top-5 shares only: critical when the largest customer holds 25% or more or the top five 60% or more, high from 15% or 45%, moderate from 8% or 30%, low below. Deeper shares and the Gini are shown without a grade.

Small books need care. With twenty customers, the top five are a quarter of the base and will hold a large share whatever the business looks like, and a Gini computed on a handful of values says little. FlowMRR does not judge concentration below 20 paying customers, while still showing the shares. Convert every customer into one reporting currency before ranking, otherwise a customer billed in a weaker currency looks smaller than they are.

Reading the result

Concentration rises for two different reasons: a large deal arrives, or smaller customers leave. The first is good news with a risk attached; the second is a retention problem that happens to show up as concentration. Check which one occurred before reacting, and re-run the figures after each large signature and at least every quarter.

Stress-test rather than only grade. Remove the largest customer, then the top three, and look at what MRR, growth and runway become. Then look at what protects those accounts: contract length, renewal date, how many people use the product inside them and whether they have payment issues. A large customer on a monthly plan with a past_due invoice is a different risk from the same customer on an annual contract renewed last month.

Frequently asked questions

What percentage of revenue from one customer is too much?

There is no universal cutoff; it depends on how committed and how stable that customer is. For reference, FlowMRR grades a largest-customer share of 15% as high and 25% as critical, and a top-five share of 45% as high and 60% as critical. Those are FlowMRR's grading rules, not industry statistics; the practical test is whether losing that customer would change your plans or your runway.

How do you calculate customer concentration?

Sum MRR per customer, sort from largest to smallest and divide each customer's MRR by total MRR. The largest customer's share and the combined share of the top five are the usual headline figures; the Gini coefficient and the Herfindahl-Hirschman index summarise the whole distribution in one number.

What is a good Gini coefficient for SaaS revenue?

There is no target value. The Gini reflects your pricing: a single-price product sits near 0, while a product combining self-serve plans and large contracts sits much higher without necessarily being riskier. Use it to follow your own trend and use the top-customer shares to judge risk.

Do investors care about customer concentration?

Yes, because it changes how predictable revenue is: losing one large account can undo months of growth. Expect questions on your largest accounts' share of revenue, their contract terms and renewal dates, and what revenue would look like without them.

Contact us
FlowMRR

Let's talk

A question, a technical issue or an idea for a partnership? Write to the FlowMRR team.