Sign in

Blog · Board and management reporting

Customer concentration: how to measure it and what the number means

Three ways to measure customer concentration from a revenue ledger, top-n share, the Herfindahl index and the count to half, what each is good for, a worked example, why the measure is reported per segment and per rep as well as for the company, and how to tell healthy concentration from dependency.

The short answerCustomer concentration is measured from the revenue ledger three ways: the share of revenue from the top n customers, usually the top 1, 5 and 10; the Herfindahl index, which is the sum of squared customer shares; and the count of customers it takes to reach half of revenue. Top-n is what a board reads, the index is what to trend, and the count-to-half is what a sales leader acts on. All three are computed per company, per segment and per rep, because concentration that is fine for the company can be total for one rep.

Every board deck has a top-ten-customers slide. Few have the same measure per segment or per rep, and fewer trend it. This guide gives three measures of customer concentration from a revenue ledger, a worked example, the roll-up, and how to separate a healthy concentration from a dependency.

Three measures

Measure Formula Use
Top-n share revenue from the largest n customers ÷ total revenue Board reporting; n = 1, 5, 10
Herfindahl index Σ (each customer's share)² Trending; sensitive to the largest
Count to half number of customers, largest first, needed to reach 50% of revenue Sales action; how many relationships carry the business

All from the same ledger, all per period.

The rows you need

  • Revenue ledger: customer, period, revenue, and rep, segment or region where the measure is to be split.

Customer identifiers only.

A worked example

Company revenue $42m across 380 customers.

Measure This year Last year
Top 1 14% 9%
Top 5 38% 31%
Top 10 52% 46%
Herfindahl 0.041 0.028
Count to half 9 13

Revenue grew. Every concentration measure rose. Four fewer customers carry half the business than a year ago. That is the finding, and the level of any single number matters less than the direction of all of them.

Per segment and per rep

The company's top customer at 14 percent is one rep's top customer at 61 percent. Split the measures:

Rep Revenue Top 1 Count to half
R-04 $6.1m 61% 1
R-11 $4.8m 18% 6
R-17 $3.9m 12% 9

Rep R-04's number is one relationship. That is a coverage and succession question the company-level figure could not raise.

The assertion

Σ customers' revenue = invoiced revenue = Σ reps = Σ segments

If the customer roll-up does not match the ledger, the shares are wrong. Check the identity before reading the measures.

Healthy or dependent

Concentrated customer Healthy signs Dependency signs
Contract Multi-year, with volume commitment None, or expiring
Sites and products Several of each One site, one product
Contacts Multiple, recent One, stale
Trend Growing Shrinking

The concentration measure says where to look. The account's own detail says what it is.

Where it goes wrong

Company level only. The rep with one customer is invisible.

Level read without trend. A top ten at 52 percent is fine in some businesses and alarming in others. The move from 46 is the finding either way.

Roll-up not checked. Duplicate customer records split a large customer into two smaller ones, and the concentration looks lower than it is.

Read as bad news by default. Concentration in the right customer is what a good sales team produces. Look before reacting.

Every quarter, at three levels

Mapped once, the revenue ledger produces all three measures per company, per segment and per rep every quarter, trended, with the roll-up checked. Covirage builds this from the export as it is. The concentration term has the short definition, and the board narrative guide covers how the number is presented with its source.

Questions people ask

What is a high customer concentration?

There is no universal threshold. Lenders and acquirers often look hard at a top customer over 10 to 15 percent, or a top ten over 50 percent, but the right comparison is the company's own history and its segment. Rising concentration with a static customer count is the signal, whatever the level.

Is concentration always bad?

No. Concentration in a growing, contracted, multi-site customer is a strong relationship. Concentration in a customer with no contract, one contact and a shrinking order is a dependency. The number says which to look at; the account view says which it is.

How is the Herfindahl index read?

Sum every customer's share of revenue squared. A single customer at 100 percent scores 1.0; a thousand equal customers score 0.001. Trend it; the level is hard to interpret alone, the direction is not.