Small business CRM audience research in England

What to take away

Define the business population, respondent role, recent CRM workflow and England rule before contacting anyone.
Treat the organisation, the buyer and the user as separate units, because one role cannot speak for the others.
Fix one definition of a small business and keep itUK government statistics use 0 to 49 employees.
Market research is not direct marketing until promotion is added or the research feeds a later sales campaign. PECR then applies.
Interview recent CRM work to understand what software needs to support, rather than relying on opinions about features.

Choose the population that can answer the question

Small business CRM audience research in England should help a firm understand what its CRM software needs to support, not just describe a broad market. Name the business size, respondent role, recent customer-record workflow and England rule before anyone is contacted; then use observed tasks and exceptions to frame software requirements.

The organisation, buyer and user are different units. A small firm may have a founder who approves contracts, a manager who owns the customer process and staff who update records. Interviewing one role cannot reveal what the others do, or give a complete picture of the software those roles need.

Research population	Decision it can inform	Evidence of eligibility
Business principal or budget holder	Whether change can be funded and authorised	Current responsibility and decision route
Process owner	Where customer work starts, changes and fails	Recent cases and who owns exceptions
CRM or spreadsheet user	What is entered, searched, corrected and exported	A task they actually did in a defined period
Privacy or contract reviewer	Which controls block or qualify a route	Role, scope and review record
Customer or prospect	How the firm's contact process feels from outside	Direct experience, not assumed buyer knowledge

Write one research question for one population. "How do CRM buyers behave?" cannot be answered as written. "How did England-based operations leads at employing firms reconcile duplicate customer records in their latest monthly close, and what would CRM software need to support that work?" names a role, a place, a task and a time. Check demand signals first to see whether the study is worth running.

Fix the definition of a small business

At the start of 2025, 5.64 million UK businesses were small (0 to 49 employees). Of all UK private sector businesses, 1.4 million (25%) had employees and 4.3 million (75%) did not employ anyone aside from the owners.

England alone had 5.0 million private sector businesses, against 361,000 in Scotland, 194,000 in Wales and 139,000 in Northern Ireland. For England the figure is 4,996,000, with a 95% confidence interval of plus or minus 117,000.

Pick one definition and keep it. A study may narrow to employers only, but it must not switch definitions once fieldwork starts.

The size counts cover the UK, so decide what England means. A workable default is the registered office in England, because Companies House registers company information and makes it available to the public. Other options are the head office, the respondent's usual workplace, the team running the CRM or the customers in the data. The market size guide shows why official counts do not identify CRM buyers.

Build the frame and the recruitment record

Build the sampling frame from records that suit the question: an authorised customer panel, a membership body's opt-in pool, a screened recruitment agency or a defined list of registered businesses. Record the frame date, coverage and known omissions. Do not present a newsletter list as the small-business population of England.

Recruitment criteria should clearly specify the people you want to take part. Agencies are generally best for recruiting the general public, so a niche role may need a different route. Write the criteria down before the first invitation goes out.

the eligible business size, sector and England rule;
the respondent role and a recent event they must have handled;
exclusions, such as suppliers, consultants or people without direct responsibility;
where contact details came from and who may use them;
invited, ineligible, declined, unreachable, withdrawn and completed counts;
access adjustments offered and accepted.

Typically between 4 and 8 participants suit a round of interviews or usability tests. That range gives depth in the recruited cases, not prevalence. A survey meant to estimate a population needs a defensible frame and statistical expertise, so compare survey tools only after the method is settled.

Keep research separate from marketing

Contacting people to conduct genuine market research is not direct marketing. It becomes direct marketing if the messages include promotional material, or if the research is ultimately carried out to send direct marketing to those people, as the ICO guidance on identifying direct marketing explains.

Direct marketing is not just about sending messages. It can include activities that lead up to, enable or support sending it, such as targeting and profiling. Screening a CRM list for a later sales push would therefore count.

The Privacy and Electronic Communications Regulations (PECR) contain the rules on direct marketing by electronic message, such as phone calls, email and text messages. Limited companies and limited liability partnerships are corporate subscribers. Sole traders and other partnerships are individual subscribers, and PECR treats them like individuals. The rule on marketing by electronic mail does not apply to corporate subscribers.

That split matters in a small-business CRM, which often holds both kinds of contact. Marketing emails or texts to individuals need specific consent. The one exception, often called the soft opt-in, covers your own previous customers. It does not apply to prospective customers or new contacts from bought-in lists.

If you collect an individual's contact details in their business capacity and intend to send direct marketing, you must tell them and have a lawful basis under the UK GDPR.

Name the data protection duties

Most processing of personal data is subject to the UK GDPR. A CRM study that records named contacts, notes or recordings is covered.

The ICO is the regulator of data protection. Organisations, including sole traders, that use personal information need to pay it a data protection fee, unless they are exempt.

Decide the lawful basis for contact details, screening answers, notes and recordings before recruitment, and have a privacy reviewer confirm it. Legitimate interests is one of the seven lawful bases for handling personal information. Consent is another, and it suits research participation best.

A research invitation must not hide lead generation. The data protection guide covers the rights requests that can follow.

Get informed consent

You need to get consent from all research participants, even if they work for your organisation. The GOV.UK page Getting informed consent lists what participants must understand, including how long their data will be kept and what their rights are.

Under the UK GDPR, consent must be a freely given, specific, informed and unambiguous indication of the person's wishes. Give participants an information sheet during recruitment. Tell them who runs the study, its purpose, what happens in the session, whether it is observed or recorded and how to withdraw. Keep any incentive log separate from responses, so payment never depends on enthusiasm for the product.

Interview recent CRM software work, not feature wish lists

Begin each interview with the participant's latest customer-record task, such as logging an enquiry or correcting a duplicate, not a feature list. Ask when the event happened, who started it, which record changed, what exception occurred, which system or workaround they used and how they knew the task was complete. This keeps software requirements grounded in work rather than hypothetical feature preferences.

Use one ledger for everything you collect, so observation never blurs into opinion.

Record type / What belongs there

Prompt: The question and any follow-up asked
Observed evidence: An action, screen or document directly seen
Participant statement: A paraphrase or approved quotation
Editorial inference: The researcher's interpretation, labelled
Unknown: A fact the session could not establish

If a participant says a CRM would save time, record a statement, not a saving. Report themes as "observed in these eligible sessions" and show the denominator. Check what interviewees say against competitor research evidence, kept as a separate class. Write customer interviews prompts from the latest enquiry rather than from a product idea. For software decisions, turn observed needs into candidate requirements: what record, task, role, exception or control the CRM must support. Keep each requirement linked to evidence and mark unresolved points for further research.

Set decision rules before fieldwork

Approve a short protocol before the first call. It should name the populations, the England rule, the frame, the recruitment route, the field dates, the interviewer, the consent materials, the analysis plan, how observations will be translated into software requirements and the deletion date.

Add thresholds. Continue when several eligible cases show the same evidenced failure. Pause if evidence comes mainly from current customers or self-selected enthusiasts. Stop if participation was presented as marketing or if withdrawal cannot be honoured.

Before you act

Name the eligible business size, sector and England rule.
Record the respondent role and a recent event they handled.
Log invited, declined, unreachable, withdrawn and completed counts.
Prepare an information sheet and a consent record.
Keep the incentive log separate from responses.
Link candidate CRM software requirements to observed evidence.
Set a deletion date for contact details and recordings.

Common questions

Short answers to the questions researchers ask most often.

Why is "busy owners" a poor audience definition?

It is too loose to recruit, analyse or challenge. It does not say which business size, which role, which workflow or which part of England the evidence covers.

Is a customer interview the same as a survey?

No. An interview explores how work happens in the recruited cases. It cannot estimate how common a behaviour is. A survey can only do that with a defensible frame, a clear design and a non-response check.

Does consent from a participant cover later sales contact?

Treat it as covering the research only. If the research is ultimately carried out to send direct marketing to the people involved, it counts as direct marketing, so keep the two purposes and records apart.