What it means
Segmentation divides a market using variables such as company size, industry, geography, buying behaviour, needs or willingness to pay. The aim is groups that are similar inside and clearly different from one another, so that a decision that makes sense for one segment does not have to be compromised for another.
It matters financially because averages hide the truth. A company reporting a blended gross margin of 30% might be earning 55% in one segment and losing money in another, and only segment-level reporting exposes that, which is why finance teams increasingly build the profit and loss account by segment rather than only by product.
The common bases are demographic or firmographic, geographic, behavioural and needs-based. Business-to-business work usually leads with firmographics such as sector, size and structure, then layers in behaviour, while consumer work leans more heavily on behaviour and underlying needs.
A usable segment has to pass four practical tests: it must be measurable, large enough to be worth serving, reachable through some channel you can afford, and different enough to justify separate treatment. Segments that fail those tests become an expensive filing exercise that changes nothing about how the business operates.
The payoff comes from directing pricing, product tiers and acquisition spend at the segments that reward them. It also sharpens forecasting, because segments usually grow at very different rates and a single blended growth assumption can be badly wrong in both directions at once.
In practice
Real-world examples
Example 1
A commercial bank segments business customers by turnover and finds that clients under $1,000,000 of revenue generate 40% of accounts but only 9% of profit. It moves that group onto a self-service digital product and reassigns relationship managers to the larger, more profitable segments.
Example 2
A gym chain segments members by usage rather than by demographics and discovers three distinct groups: daily attenders, twice-weekly classes and rarely-attending direct debit holders. Each group gets a different retention approach, and overall annual churn falls by roughly a fifth.
Example 3
A packaging supplier splits customers by order pattern and finds that small, frequent orders cost far more to serve than their price reflects. It introduces a minimum order value for that segment, losing a handful of accounts but improving contribution across the remainder.
Think of it
Market segmentation is dividing customers into groups so you can serve each group's specific needs.
Formula
Calculation
Segment Revenue Potential = Segment Size x Expected Penetration x Average Revenue per Customer
Segment Contribution = Segment Revenue - Segment Variable Costs - Segment-Specific Fixed Costs
A business software company evaluates a segment of 250,000 small retailers and expects to reach 4% of them within three years.
Customers = 250,000 x 0.04 = 10,000.
At an average revenue per customer of $1,200 a year:
Segment revenue = 10,000 x $1,200 = $12,000,000.
Variable costs, mainly hosting, payment processing and support, run at 40% of revenue: $12,000,000 x 0.40 = $4,800,000.
Segment-specific fixed costs, covering a dedicated account team and sector marketing, are $2,200,000.
Segment contribution = $12,000,000 - $4,800,000 - $2,200,000 = $5,000,000.
Contribution margin = $5,000,000 / $12,000,000 x 100 = 41.7%. That figure can then be compared directly with other candidate segments to decide where the next hire and the next marketing dollar should go.Case study
Seen in the real world
This is an illustrative and fictional example. Wrenfield Logistics, an invented regional freight company, priced almost entirely on distance and weight and reported a steady 12% operating margin across a $38,000,000 revenue base. Management assumed profitability was fairly even across the customer base.
A segmentation exercise grouped customers by delivery density, booking lead time and returns rate rather than by size. Three segments emerged. Dense urban retail customers with long lead times generated a 26% contribution margin, scheduled industrial customers came in at 14%, and last-minute e-commerce customers with high returns produced a contribution margin of just 1% while consuming a disproportionate share of driver hours.
Wrenfield built a surcharge for short-notice bookings, introduced a returns handling fee, and gave its sales team a target weighted towards the dense urban segment. Two years later revenue was slightly lower at $36,500,000 but operating margin had risen to 17%. The illustrative lesson is that segmentation is most valuable when the segments are defined by what actually drives cost, not by what is easiest to measure.
Watch out
Common mistakes
- Segmenting by characteristics that do not change any decision. If two groups get the same price, the same product and the same marketing, splitting them apart adds work without adding value.
- Creating too many segments. A dozen finely sliced groups are impossible to serve distinctly with a small team, and most companies do better with three to five that they act on properly.
- Segmenting customers without also allocating costs to them. Revenue by segment alone can point in exactly the wrong direction if the biggest segment is also the most expensive to serve.
Questions
People also ask
How many segments should a business have?
Usually between three and six, chosen so that each is large enough to justify its own approach and the team can genuinely deliver different treatment to each.
What is the difference between segmentation and targeting?
Segmentation is dividing the market into groups, while targeting is the separate decision about which of those groups you will actually pursue and invest behind.
How often should segmentation be revisited?
Every two to three years in most markets, or sooner if buying behaviour shifts, a new channel emerges, or actual profitability starts diverging from what the model predicted.