B2B market research: how to study customers that are companies, not people

1. The problem: market decisions based on what the sales team hears

A corrugated packaging manufacturer has a few hundred customers, twenty of whom generate most of its sales. The commercial director knows what the large clients say at annual meetings and what the sales reps report about competitors. He does not know why mid-sized customers are ordering less and less often, which criteria purchasing departments use when they compare suppliers, or whether a recyclable packaging line would actually be paid for or merely requested in meetings. Decisions are made on what reaches the team's ears, and what reaches them is filtered through personal relationships and individual interests. B2B market research exists to replace those impressions with a systematic picture of the business market.

The problem is not specific to one industry. A software firm selling to banks, a medical equipment distributor, an accounting services provider for small firms, a business association that wants to know which services its members expect: all of them have customers that are organisations, and organisations buy differently from people.

2. What B2B market research is and how it differs from consumer studies

Business-to-business research applies the same principles as any market research: a clear question, a suitable sample, a valid instrument, rigorous analysis. What differs is the nature of the market under study, and those differences change almost every design decision. Five are essential.

• Several people decide. An equipment purchase involves the user, the engineer who writes the specification, the purchasing department, the finance director and sometimes senior management. Each sees different criteria.
• Buyers are professionals. They compare offers systematically, know the market and quickly spot a question written by someone who does not understand the field.
• The process is long. Months can pass between the need arising and delivery, with stages of specification, tendering and negotiation.
• Every customer matters. A handful of clients can account for a large part of turnover, so losing a single one is a strategic problem.
• Purchases are not isolated events. They are part of a long-term relationship, with a history, trust and the cost of switching supplier.

These differences produce the four practical challenges of any B2B study: sampling, access to respondents, respondents' memory of long processes, and capturing the relationships between organisations.

3. When it is the right choice and when it is not

A B2B study is the right choice when the company is entering a new customer segment or a new market, when it is developing a product or service and needs to learn which purchase criteria matter, when it wants to measure customer satisfaction and loyalty comparably from year to year, when it is losing customers and does not understand why, or when it needs to estimate what customers would pay for a new offer, a subject covered in the guide on pricing research.

It is not the right choice when the company has five customers whom management knows personally: structured relationship reviews, conducted by the managers themselves, will yield more than a formal study. Nor is it suitable when the question concerns the market as a whole and the answer already exists in official statistics, company registers, industry reports and foreign trade data; there, the first stage is desk research, and fieldwork covers only what the sources cannot.

4. The sample: small and highly unequal populations

In a consumer study the population runs to hundreds of thousands or millions of people, each of roughly equal importance. In B2B market research the population may be a few hundred or a few thousand companies, and the differences between them are enormous: a retail chain with hundreds of staff and a corner shop are, statistically, two equal units, but commercially they have nothing in common.

The sample is therefore built differently. The sampling frame is assembled from company registers, commercial databases, trade association lists and, for studies of the company's own customers, its customer database. Every source has gaps, and combining them requires cleaning and de-duplication. Stratification is by size, usually headcount or turnover, and by sector. Large companies, few and important, are included in full or almost in full; small companies are sampled. At the analysis stage, results can be reported in two ways that must be kept explicitly apart: weighted by the number of companies, which shows what a typical firm does, or weighted by size or purchase volume, which shows where the market actually is.

5. The right respondent: the role in the buying centre

The unit being studied is the company, but someone has to answer on its behalf. Choosing that person is perhaps the most important decision in the study. Marketing literature describes an organisation's buying centre in terms of roles: users, who work with the product; influencers, who write the specifications; buyers, who negotiate; deciders, who approve; and gatekeepers, who control the flow of information. For a question about technical quality, the right respondent is the user or the engineer; for a question about price and terms, the person in purchasing; for a strategic question, the director.

The questionnaire therefore opens with screening on role: "are you directly involved in choosing suppliers of...?", "what is your role in this decision?". In major studies, several members of the same company's buying centre are interviewed, which shows where their views diverge, often the most useful finding for the sales team.

6. Methods: what works in business markets

In-depth interviews with decision-makers are the dominant method. They allow complex processes, real selection criteria and the relationship with suppliers to be discussed in the respondent's own language and at a time of their choosing. The guide on in-depth interviews with experts and decision-makers explains how they are prepared and conducted, and the focus groups and IDI service runs them in the field.

Telephone surveys work for larger populations, such as small and medium-sized firms, with appointments booked in advance and a 10-15 minute questionnaire. CATI makes it possible to reach the right person after several attempts and to keep strict control of screening. Online surveys are efficient where named email addresses exist and respondents are used to working at a computer, but response rates are low without a personalised invitation and a clear reason to take part. Focus groups are rare in B2B: competitors do not speak openly in the same room, and decision-makers are hard to gather at the same hour.

Access and credibility are, in practice, the limiting factor. Decision-makers receive many requests, and an assistant screens the calls. What helps is an invitation letter that explains the purpose and guarantees confidentiality, an interviewer who understands the field and can follow a technical discussion without reading the questions mechanically, flexible scheduling, and some form of reciprocity, for example a summary of the aggregated findings.

7. Analysis: small bases, big decisions

Analysis in a B2B study combines figures with interpretation. On the quantitative side, results are presented by size stratum and by sector, with the weighting stated explicitly, and differences are tested statistically only where the bases allow. Analysis of the factors driving loyalty and supplier choice, similar to that described in the guide on customer satisfaction and NPS, works if the sample holds at least one or two hundred companies.

On the qualitative side, interviews with decision-makers are analysed thematically: selection criteria, the moments when the supplier won or lost a contract, perceptions of competitors, future needs. A hypothetical example: interviews with the packaging manufacturer's mid-sized customers show that orders are falling not because of price, but because delivery dates are not confirmed in writing and the customers' production departments cannot plan their lines. The fix is operational, not commercial, and would never have surfaced from sales reports.

8. Limits and common errors

• The wrong respondent. Interviewing whoever was available rather than whoever decides produces consistent data about something other than the question.
• Memory of long processes. Respondents reconstruct decisions made months ago, often rationalising them. Questions about the most recent concrete purchase reduce the distortion.
• Percentages on small bases. "40% of customers" calculated on 15 companies is not a finding; it is six companies.
• Confused weighting. Results weighted by number of firms and results weighted by volume say different things; mixing them misleads.
• Ethical ambiguity. Gathering intelligence on competitors under the pretext of a study is incompatible with professional codes, including the ICC/ESOMAR International Code. Legitimate B2B market research introduces itself honestly, guarantees respondents' confidentiality and does not pass identifiable answers to the client without consent.
• Over-surveying. The same large companies are interviewed often, and their fatigue lowers the quality of answers.

9. Consumer studies and B2B research compared

10. What to ask the agency

• The source of the sampling frame, how it is cleaned and what it covers.
• Stratification by size and sector, and how large companies are handled.
• The definition of the right respondent and the role-screening questions.
• The method for each segment of companies, with the reasoning.
• The profile and briefing of interviewers for the field in question.
• The access strategy: invitation letter, appointments, reciprocity.
• Confidentiality rules towards respondents and for reporting company-level data, in line with the GDPR and Moldova's Law no. 133/2011 on personal data protection.
• How weighting will be applied and small bases reported.

Business markets are smaller than consumer markets, but every wrong decision costs more. Well-designed B2B market research gives sales and management a picture of customers that does not depend on who has the best relationship with whom. If you are preparing to enter a new segment or want to understand why you are losing customers, the Perspicuus qualitative research and data analysis team can help you build the right study.

See also: Focus groups and IDI CAPI and CATI Data analysis and desk research In-depth interviews with experts and decision-makers Desk research: what you learn before fieldwork Customer satisfaction and NPS

The Perspicuus research team · 9 September 2025