Customer concentration: share of the largest
From a list of customers and their revenue computes the share of the largest, the top N and all the rest, the concentration index and the effective number of customers, and what happens to revenue and contribution if part of the top-N revenue leaves.
At a glance
- What question does it answer?
- How far revenue depends on a few large customers and what happens if part of their revenue leaves.
- What data is needed?
- A list of customers with revenue for one period excl. VAT; small customers can be merged in one row with their number.
- How to read the result?
- On the example 10 customers with total revenue of 100m. The largest gives 42% and the three largest 71%. The concentration index is 2,364, effectively 4.2 equal customers. If half of the three largest customers’ revenue leaves, the loss is 35.5m (35.5%) of revenue and 12.4m of contribution at a contribution margin of 35%. Example on the default values: Share of the largest customer: 42%; Share of the 3 largest: 71%; Share of all the rest: 29%; Concentration index (sum of squared shares × 10,000): 2,364. The figures are illustrative: enter your own to see your result.
Why and how to reason
If one customer brings 42% of revenue, its loss is not “one of ten” but a blow to the whole business. To see it in numbers we count not only the share of the largest but also the concentration index: the sum of squared shares × 10,000 (Herfindahl index, HHI). Squaring gives larger customers more weight: in the example 42² = 1,764 of the total 2,364. From the index we get the “effective number of equal customers”: 10,000 ÷ 2,364 = 4.23. In risk terms ten customers equal just over four identical ones. The scenario turns it into money: the top 3 hold 71% of revenue, 71 million; if half leaves, we lose 35.5 million of revenue and, at a 35% contribution, another 12.4 million of contribution. For those who pay fixed costs the contribution is what matters: it disappears while costs stay.
When to use it
It comes in handy in B2B: a brand with several wholesale buyers, a supplier to chains, a contract manufacturer. Useful when valuing a business and planning a reserve. It does not fit mass retail with thousands of small buyers where the index is always small. Assumptions: revenue per customer for one period; “n customers” rows split revenue equally and the index is then a lower estimate. There is no “dangerous” concentration threshold in Russia: set your own.
How to read the result
First the largest customer’s share: compare with your own limit (say 25%). Then the top N and “all the rest”: if the rest give less than 30%, the business hangs on a few contracts. The effective number of customers is easier to explain to management than the index. The scenario loss of contribution shows how many fixed costs would be left uncovered and helps decide whether a reserve or diversification is needed. Watch not the absolute but the index trend from period to period: growth means the business is becoming dependent.
Common mistakes
Counting customers by legal entity when it is really one group of companies: concentration is understated. Taking revenue excluding VAT in one row and including in another. Using share of revenue rather than of contribution: a low-margin customer is less dangerous than a high-margin one.
What data is needed?
- A list of customers with revenue for one period excl. VAT; small customers can be merged in one row with their number.
- How many of the largest to count, the share leaving and the contribution margin.
Inputs
| Field | Unit | Value in the example | Meaning |
|---|---|---|---|
| How many of the largest to count | pcs | 3 | How many of the largest customers to include in the concentration measure (top N). |
| Scenario: what part of top-N revenue leaves | % | 50 | Scenario: what part of the top-N revenue leaves, in percent. Shows what happens to revenue if large customers are lost. |
| Contribution margin | % | 35 | Optional. The share of revenue after variable costs. |
| Your limit for one customer’s share | % | — | There is no single standard in Russia: enter your own figure or leave it empty. |
How it is calculated
| Step | Formula |
|---|---|
| Customer share | customer revenue ÷ revenue of all (in a row of m customers: row revenue ÷ m ÷ revenue of all) |
| Index | sum (share²) × 10,000 over individual customers |
| Effective number | 10,000 ÷ index |
| Loss | top-N revenue × share leaving × contribution margin |
Worked numbers
Default values from the app; in the interactive version you enter your own.
- Share of the largest customer
- 42 %
- Share of the 3 largest
- 71 %
- Share of all the rest
- 29 %
- Concentration index (sum of squared shares × 10,000)
- 2,364Computed by individual customers; a row with a number of customers is split into equal shares.
- Effective number of equal customers
- 4.210,000 ÷ index: with equal customers it equals their number.
- Scenario: lost revenue
- 35.5 m ₽ · 35.5 %
- Scenario: lost contribution
- 12.4 m ₽
Customers in descending order
| Customer | Customers | Revenue | Row share | Cumulative share |
|---|---|---|---|---|
| Customer 1 | 1 | 42 m ₽ | 42 % | 42 % |
| Customer 2 | 1 | 18 m ₽ | 18 % | 60 % |
| Customer 3 | 1 | 11 m ₽ | 11 % | 71 % |
| Customer 4 | 1 | 8 m ₽ | 8 % | 79 % |
| Customer 5 | 1 | 6 m ₽ | 6 % | 85 % |
| Customer 6 | 1 | 5 m ₽ | 5 % | 90 % |
| Customer 7 | 1 | 4 m ₽ | 4 % | 94 % |
| Customer 8 | 1 | 3 m ₽ | 3 % | 97 % |
| Customer 9 | 1 | 2 m ₽ | 2 % | 99 % |
| Customer 10 | 1 | 1 m ₽ | 1 % | 100 % |
There is no “dangerous” concentration threshold in Russia: set your own by how much revenue you can afford to lose with one customer. Watch how the index changes from period to period.
Nuances and limits
- There is no “dangerous” share threshold in Russia: set your own.
- Merge customers of one owner group. Merge small customers only with their number in the “Customers in the row” column: otherwise the row counts as one large customer and the index is overstated.
- Track the index by period, not on one date.