Which customer differences change how you sell?

Start with the commercial decision. You may need different material for a first-time project buyer and a customer replenishing the same parts. Or you may need to distinguish accounts that require technical review from those that can order a supported item directly. The segment should explain what your team would do differently.

Company size, industry and location can help describe a group, but a label alone is not the reason to split it. Two manufacturers of similar size may buy in very different ways. Two industries may share the same maintenance requirement and buying process.

Write a test sentence: “We would handle these accounts differently because…” Complete it with the buyer’s need or an operating condition. If the only change is a database tag or the logo on a presentation, the proposed split may not yet be useful.

Set the scope before comparing accounts: one product family, service region or commercial question. Use the ideal customer profile for required fit and exclusions. Segmentation then distinguishes suitable customers; it should not hide accounts the business cannot serve.

Select accounts with different buying and delivery experiences. Review the problem they brought, how they evaluated the offer, who was involved, the work needed before purchase and the support required afterwards. Use permitted account notes, order history and conversations with the people doing the work.

Ask customers about the decision rather than inviting agreement with your segment names. “What had to be resolved before this order could proceed?” can reveal a useful distinction. “Are you a strategic project buyer?” mainly tests whether they recognise your terminology.

The Netherlands Chamber of Commerce (KVK) publishes target-customer research guidance that encourages using existing customer information and asking what mattered in the buying choice. Keep a record of which accounts support your proposed difference and which do not.

Include the cost and constraints of serving the work. A project may need drawings, site coordination and several approval rounds. Repeat orders may need accurate stock information and dependable exception handling. Revenue alone can conceal those different demands on the team.

Compare needs, buying conditions and service work together. Useful differences might include repeat versus project purchasing, standard versus engineered requirements, or local versus central approval. Explain the evidence behind each distinction rather than choosing whichever fields are easiest to export.

Avoid multiplying segments for every combination. A delivery region may be an ownership rule that applies across both repeat and project work. A supported product category may change the specialist involved without requiring a separate campaign. Keep a distinction at the level where it changes a decision.

A single account can contain more than one buying situation. A plant may reorder standard consumables while procuring equipment for an expansion. You can segment the buying task or site instead of forcing the entire group into one permanent label.

Record overlap and unknowns explicitly. For an unclassified account, record the question that would resolve its segment, such as whether the proposed purchase needs technical approval. It should not receive a segment merely because a blank value would complicate a filter.

Write a short rule for each segment, the evidence needed and an unresolved state. For example, “project purchasing” could require a specified installation or changed technical requirement that needs review. An unusually large order is a clue to investigate, not sufficient evidence by itself.

Ask two colleagues to assign a mixed set of accounts or buying situations independently. Compare disagreements about the unit, evidence and treatment. If a colleague cannot tell which materials or handover should change after assigning the segment, refine the rule before scaling it.

Keep the segment beside its date, source and purpose in the account list. A label created for a campaign should not silently determine service priority or credit decisions. Those uses need their own rules and owners.

Test a boundary case as well as a clear example: a repeat customer starting a new project, a central buyer ordering for several sites, or a project that returns to routine replenishment. Decide who updates the record and what existing work must remain with its owner.

Use the distinction to prepare a relevant answer. A repeat-order customer may need accurate availability and a simple way to resolve a mismatch. A project buyer may need to see how technical uncertainties and approvals are handled. The positioning and messaging guide connects those questions to supported claims.

Carry the same segment logic into the destination of a campaign. An advertisement promising help with a specification should lead to material that answers that task, and an enquiry should reach someone able to review it. The paid-campaign brief covers the audience, message and landing-page handover.

Use segment knowledge with current customers too. A routine account entering project work may need an account conversation. The whitespace guide helps distinguish a potential gap from a category the customer has no reason to buy.

The channel is a decision to test, not a property of the segment. A project buyer may arrive through a referral, a salesperson or useful technical content. Choose the channels the evidence and available team can support.

Compare the proposed treatment with what actually happened. Did the material answer the question? Could the recipient handle the request? Which accounts required different treatment from the assigned segment? Record the exceptions and the reason for each change.

Look for evidence that challenges the split. If both groups repeatedly need the same explanation, one shared asset may be enough. If apparently similar project buyers have different approval requirements, a more specific buying condition may be useful. Do not judge the segmentation merely by whether every account has a label.

Keep outcomes and denominators consistent. A higher conversion rate in one small segment may reflect existing relationships or a different stage of buying. It does not show that the new segmentation caused the result.

Agree a review trigger: a new offer, changed delivery model, enough unresolved assignments to disrupt use, or a scheduled account review. Stable rules are easier to operate than categories that change whenever one unusual deal appears.

Choose one offer and write two candidate customer groups. For each, record the need, buying condition, evidence, changed treatment and expected owner. Add one account that does not fit neatly and explain how you would handle it. Then ask a colleague to use the result to prepare a real piece of work.

Commission the investigation independently through Plan, with findings your team can use. When the groups and approach are already clear, build on that work. An agreed campaign can cover the relevant accounts, messages, outreach and response handling, with responsibilities divided to suit your team. A company-wide segmentation programme is not a prerequisite.