Market segmentation divides a market into groups of buyers who want different things and would be served well by different products or different marketing. It is the first of the three decisions that marketing strategy is built on, and it is the one most B2B companies think they have already done.

Most have not. They have a list of industries and company sizes, taken from the fields their CRM happens to hold, applied to the customers they already have. That is a description of a customer base, not a segmentation of a market.

A customer base tells you who already buys from you. A segmentation tells you who is in the market, which is what targeting and positioning need in order to work.

What Segmentation Is For

Kotler and Armstrong define market segmentation as dividing a market into smaller groups of buyers with distinct needs, characteristics or behaviours who might require separate products or marketing mixes.

The purpose is to make a market you cannot address as a whole into something you can make decisions about. A market of forty thousand companies is not something a marketing team can act on. Five groups within it, each wanting a different thing for a different reason, is.

Segmentation is about the market, not about the company. It describes buyers who exist whether or not you sell to them. A competitor working from the same evidence should arrive at roughly the same segments.

A segmentation also covers the whole market. Every buyer in the category belongs to exactly one segment, including the buyers who currently use a competitor and the ones nobody is serving. Leaving them out means the segmentation cannot show you where new customers would come from.

Deciding which of those segments to serve comes next. That decision is targeting, and it is where the company commits and where the trade-offs are.

Firmographics Are Usually Not Segments

Industry, headcount, revenue band, region and tech stack are the easiest lines to draw in B2B, because they are the fields the CRM already holds. Groups drawn on those lines usually contain buyers who want different things.

A 400-person manufacturer and a 400-person law firm sit in the same headcount band. One is buying to reduce downtime on a production line, the other to reduce the hours partners spend on administration. The same proposition will not work on both, and the same media will not reach the people who decide.

If two groups would be served well by the same product and the same marketing, they are not separate segments, however different the labels look. The reverse also holds. If two companies in different industries and different size bands want the same outcome for the same reason and buy the same way, they are in one segment, and splitting them costs you scale for nothing.

This is not an argument for throwing firmographics away. They do a job, and it is a job further down the process. Once a segment is defined by what its buyers need, firmographics are how you find those buyers, size the segment and buy media against it. The mistake is starting there.

What a Segment Is Actually Built From

A segment holds together when its members share what they are trying to achieve, what they believe about the category, and how they buy. What a buyer is trying to achieve changes what they buy and who they buy it from. Headcount does not.

In B2B that usually means asking what problem triggers the purchase, what the buyer is measured on, what happens if they do nothing, who else has to approve the decision, and what would make them switch away from what they use today. Two companies that answer those questions the same way will respond to the same proposition, even if one is a hospital and the other is a logistics firm.

Firmographics come back in at the end. Once the groups exist, you look at what they have in common that is visible from outside: the industries they cluster in, the sizes they tend to be, the roles that lead the buying. Without those visible characteristics you cannot buy media against the segment or tell the sales team who to call.

None of this is available from the CRM. It comes from research: interviews and conversations to find out what the issues are, then a survey with a representative sample to find out how common each issue is and how large each group is. Qualitative research alone tells you what exists without telling you how much of it there is. Quantitative research alone measures answers to questions you may have framed wrongly. The two together are what produces a segmentation you can size.

A B2B Market Segmentation Example

Take a company that sells maintenance management software, which organisations use to schedule repairs and servicing on their equipment. The market is every organisation that maintains equipment, whether it uses this company's software, a competitor's software or a spreadsheet. In this example the market holds 20,000 organisations, and research shows they divide into four groups by what they're trying to achieve.

Example figures for illustration.

Industry doesn't decide which segment an organisation is in. A hospital that has just failed an inspection belongs in Audit Ready. A hospital group that has just bought three other hospitals belongs in One System. Both are hospitals, but they want different things, different people lead the purchase, and a different message will work on each.

Firmographics still have a job. The marketing team uses the "Where you find them" column to buy media, and the sales team uses it to build account lists. That column is filled in last, after the segments have been defined by what buyers need.

The Requirements a Usable Segmentation Has to Meet

Kotler and Armstrong set five requirements for effective segmentation. A segmentation that fails one of them usually looks fine until someone tries to plan a campaign from it.

Measurable. The size and purchasing power of each segment can be measured. If you cannot say how many companies are in a segment and roughly what the category is worth to them, you cannot compare it against the others, and targeting becomes a matter of preference.

Accessible. The segment can be reached and served. A group defined purely by attitude often fails here. If the only thing its members have in common is a belief, there is no way to find them.

Substantial. The segment is large enough to be worth serving. In B2B this catches the segment built around three flagship accounts. Three accounts are a sales plan, not a segment.

Differentiable. The segments respond differently to different marketing. This is the requirement firmographic segmentations usually fail. If two segments would respond the same way to the same campaign, you have split one segment in two for no reason.

Actionable. The company can build programmes to attract and serve the segment. If the product cannot serve a segment and the sales team cannot support it, the company cannot target it this year. Record it and move on.

How to Size a B2B Market Segment

You size a segment by counting the organisations in it and estimating what the category is worth to them each year. Without both numbers you can't compare segments, and the targeting decision comes down to preference.

Count the whole market first. Use sources that cover every organisation with the problem, not only your customers: government business statistics, industry and trade association data, and regulatory registers. In the example, the count is 20,000 organisations that maintain equipment.

Work out how many organisations sit in each segment. Survey a representative sample of the market using the questions that define the segments. Then apply the share of respondents in each segment to the market size. If 20 per cent of respondents fit Downtime Prevention, that segment holds about 4,000 of the 20,000 organisations.

Estimate the value of each segment. Multiply the number of organisations by what a typical one spends on the category each year. In the example, Downtime Prevention is 4,000 organisations spending about $30,000 a year, which is $120 million. Off the Spreadsheet is 9,000 organisations spending about $3,000 a year, which is $27 million. The segment with the most organisations is the smaller market by value. These are example figures for illustration.

Size alone doesn't decide which segment to serve. A large segment is still out of reach if the product can't serve it or the sales team can't support it. That decision belongs to targeting, which B2B Market Targeting: How to Choose Which Segments to Serve explains.

How to Tell Whether Yours Is Real

Does it include buyers you do not have? If every segment is populated entirely by current customers, the work described the customer base rather than the market, and it cannot show you where the growth is.

Can every buyer be placed in exactly one segment? If a company could sit in two segments, two teams will market to it in two different ways.

Is each segment named for what its buyers want? A segment called Mid-Market Manufacturing tells you where to find people. A segment named for the outcome its buyers want tells you what to say to them.

Can the sales team name real companies in each one? If the people who talk to buyers every day cannot place accounts into the segments, the segmentation was built from data rather than from the market, and nobody will use it.

Where Segmentation Stops

A finished segmentation makes no commitments. It says the market divides into five groups, here is what each one wants, here is how big each one is, and here is where each one buys today. Nobody has chosen anything yet.

Targeting is the next decision, and it is about the company rather than the market. It commits the product roadmap, the budget and the sales team, so it should be made separately, by the people accountable for those commitments.

One warning about what happens next. Choosing which segments to serve is not the same as choosing how narrowly to advertise. The evidence in B2B points hard against narrowing reach inside a chosen market, and the hub article on segmentation, targeting and positioning sets out what the research shows.

FAQ

Is customer segmentation the same as market segmentation? No. Customer segmentation divides the companies that already buy from you, usually to decide how to serve them or what to charge them. Market segmentation divides every buyer in the market, including companies that buy from a competitor and companies that buy nothing today. A segmentation built only from your customers can't show you where new customers will come from. Your own sales data becomes useful at the targeting decision, where won and lost deals show which segments you can win.

What variables can you segment a B2B market by? Benson Shapiro and Thomas Bonoma set out five layers of variables in Harvard Business Review in 1984. The outer layer is demographics: industry, company size and location. Next come operating variables, such as the technology a company uses and its capabilities. Next comes the purchasing approach: how buying is organised, the buying criteria and the supplier relationships. Next come situational factors, such as urgency, how the product will be used and order size. The inner layer is the personal characteristics of the buyers. Shapiro and Bonoma recommend starting at the outer layer and moving inwards until the segmentation is good enough, because each layer inward costs more to research. Segments defined by the outer layer alone usually fail the differentiable test, because buyer needs and buying behaviour sit in the inner layers.

How many segments should a B2B market have? There's no fixed number. Each segment has to want something different and be large enough to be worth serving. If two segments would get the same product and the same marketing, merge them. If one segment contains buyers who want different things, split it.

What's the difference between segmentation and targeting? Segmentation describes the market: which groups of buyers exist, what each group wants and how large each group is. Targeting is the decision about which of those segments the company will serve. Segmentation makes no commitments. Targeting commits the product, the budget and the sales team.

Sources

• Philip Kotler and Gary Armstrong, Principles of Marketing. Source of the definition of market segmentation, and of the five requirements for effective segmentation: measurable, accessible, substantial, differentiable and actionable.
• Benson P. Shapiro and Thomas V. Bonoma, "How to Segment Industrial Markets", Harvard Business Review, May to June 1984. Source of the five layers of B2B segmentation variables and the advice to work from the outer layer inwards.