Revenue concentration analysis

Revenue concentration analysis measures how much of total revenue depends on a small number of customers, products, or segments. The plainest version answers one question: if your biggest account walked tomorrow, how much of the business goes with it. A company where the top five customers produce 60% of revenue is in a very different risk position from one where the top five produce 15%, even if both grow at the same rate.

The metric matters because revenue that looks healthy in aggregate can be dangerously fragile underneath. Strong total growth driven by two enterprise accounts hides the fact that a single renewal decision can erase a year of progress. Investors, boards, and acquirers scrutinise concentration precisely because it is the gap between a number that is large and a number that is durable. It sits alongside net revenue retention as a test of revenue quality rather than revenue size.

Concentration is not automatically bad. Early-stage companies almost always start concentrated, and deep relationships with a few large customers can be a moat. The risk is being concentrated without knowing it, or staying concentrated long after the business should have diversified. Analysis turns a vague sense of dependence into a tracked number you can manage deliberately.

The simplest measure is top-N concentration: sum the revenue of your largest N customers, divide by total revenue, and multiply by 100. If your top 10 customers generate 480,000 pounds out of 1,200,000 pounds in total revenue, your top-10 concentration is 40%. Common cuts are top 1, top 5, top 10, and top 20% of customers.

For a single summary number, many teams use the Herfindahl-Hirschman Index, which squares each customer share and sums the results. A perfectly even spread across many customers produces a low index; a business with one dominant account produces a high one. Whichever measure you use, run it across more than one dimension, because customer concentration, product concentration, and industry concentration are distinct risks that can each sink you independently.

1. 1

   Pick the dimension you are testing

   Customer, product, industry, or geography. Each is a separate concentration risk. A diversified customer base can still be dangerously concentrated in one vertical that moves together.

2. 2

   Choose the revenue basis

   Recurring revenue, total bookings, or gross profit. Recurring revenue is usually the cleanest basis for concentration risk because it is what you stand to lose at renewal.

3. 3

   Rank entities and sum the top N

   Order customers from largest to smallest and add up the revenue of the top 1, 5, and 10 so you can see how steeply the curve falls off.

4. 4

   Divide by total revenue

   Express each top-N total as a percentage of total revenue. Track these shares over time, since the trend reveals whether you are diversifying or quietly concentrating further.

5. 5

   Add a single-number index if useful

   Compute the Herfindahl-Hirschman Index for a board-friendly summary that captures the whole distribution rather than just the top slice.

Concentration is an outcome of forces you can name and manage. It rises when a few accounts expand fast, when the long tail churns, when new logos arrive too slowly to dilute the leaders, and when whole segments move in lockstep. A metric tree separates these forces so a worsening concentration number points at a specific cause rather than a general unease.

Decomposing it this way also makes the fix ownable. If concentration is climbing because new logo acquisition has stalled, that is a sales and marketing problem. If it is climbing because the long tail is churning, that is a retention problem tied to churn rate. Different leaves, different teams, different actions.

Benchmarks here describe risk thresholds rather than targets, and they shift with stage. A seed-stage company living off three design partners is concentrated by design, while a public company with the same profile would be flagged immediately. The ranges below are the levels at which most boards and acquirers start asking harder questions.

Improving concentration usually means lowering it, but the goal is managed exposure rather than a number for its own sake. You reduce risk by broadening the base, protecting the accounts you depend on, and seeing the danger early enough to act.

1. 1

   Measuring only customer concentration

   A spread of customers all in one industry, region, or product line is still concentrated. Looking at a single dimension creates false comfort about a base that moves together.

2. 2

   Treating high concentration as always bad

   Deep dependence on a few customers can be a deliberate, defensible strategy early on. The failure is unmanaged concentration, not concentration itself.

3. 3

   Watching the level but not the trend

   A 30 percent top-5 share that is falling is a very different story from a 30 percent share that is climbing. The direction carries more information than the snapshot.

4. 4

   Forgetting concentration on the cost side

   Dependence on a single supplier, channel, or platform is the mirror image of customer concentration. Revenue diversity means little if one vendor can switch you off.