Seven steps to stronger professional services marketing audience research

What to take away

Most audience research in professional services fails before a single interview is booked. A firm decides it needs better insight, commissions a survey, and then finds the questions do not map to any decision the partners are willing to make. The report is filed. Nothing changes.

The fix is sequence. Decide what the research must inform, audit what you already know, then design the smallest piece of fieldwork that closes the gap. This guide sets out that sequence in seven steps, with the UK rules and data sources that apply.

Step one: name the decision the research must serve

Write the decision down in one sentence. For example: "Should we launch a fixed-fee employment retainer for small employers in the North West?" That sentence tells you who to research, what to ask and when to stop.

Express it as something you could answer yes or no to. "Is there demand for this service at this price among this group?" is testable. "How do clients feel about US?" is not, and it will produce a report nobody can act on.

Research without a decision attached drifts towards interesting but unusable findings. Partners rarely fund interesting. They fund evidence that reduces risk on a choice they are about to make.

Assign one owner to the decision before fieldwork starts. Work owned by a committee stalls. Work owned by one person with a deadline tends to finish.

If you cannot name the decision, you are not ready to research. Go back to the business plan. A useful prompt is to ask which budget line would change if the findings came back the opposite way.

Step two: audit the evidence you already hold

Most firms hold more evidence than they realise. It sits in inboxes, CRM notes and the heads of partners. Pull it into one place and tag each item by source, date and theme.

Where two sources disagree, you have found something worth testing in the interviews. Note it rather than resolving it early.

This audit also shows you what is stale. A view of the market formed in 2019 may still be shaping your positioning. Check it against current conditions, including cost pressures. The ONS consumer price inflation time series is the reference dataset for how prices have moved, which matters when you test fee sensitivity.

Step three: map the buying group, not just the buyer

Professional services purchases usually involve several people. A finance director may sign, a managing partner may influence, and an operations lead may do the searching. Each has a different question.

Map those roles for your three largest client types. Note who researches, who shortlists and who vetoes. This map becomes the spine of your interview guide and your messaging.

Map the sequence as well as the roles. The person who first searches for a firm may never appear in your enquiry log, and the person who signs may never read your website. Interview the searcher.

Our guide to professional services marketing target audience covers how to turn that map into a defined, reachable group rather than a demographic guess.

Step four: choose methods that fit the decision

Match the method to the question. Interviews suit questions about language and process, while surveys size how common a preference is. Behaviour is better read from analytics and enquiry records.

Qualitative work first is usually right. Ten to fifteen conversations will surface the vocabulary and the objections. You then test those in a wider survey.

Run the qualitative phase before you write survey questions, not alongside it. Questions written in advance of interviews tend to test the firm's assumptions rather than the client's reality.

Budget fieldwork in hours, not pounds, when the work is internal. Fifteen interviews at 45 minutes each, plus recruitment and analysis, is roughly a fortnight of one person's time.

Our guide to professional services marketing survey tools compares question logic, data exports and retention settings, which matter more than the template a vendor supplies.

Step five: build the sample without fooling yourself

Sample quality decides whether the findings are usable. A survey of your own contact list tells you about your contact list, not the market. Say so in the report.

For market-level claims, recruit beyond existing clients. Include people who chose a competitor and people who chose to do nothing. The second group is the one most firms ignore and the one that explains lost revenue.

Record how each participant was recruited and what you offered them. That record is also your audit trail if anyone later questions the basis on which you hold their details.

Incentives lift response rates and introduce bias at the same time. Note what you offered, and consider a summary of findings instead of a voucher for client panels.

Step six: run fieldwork within the UK GDPR rules

Before you collect personal data, identify your lawful basis and write it down. The ICO's guidance on lawful basis under the UK GDPR sets out the six options, including legitimate interests and consent.

Consent is not automatically the right choice. It can be withdrawn, which makes it a poor fit for long-running research panels. Legitimate interests often fits market research better, provided you complete a balancing test.

Keep the principles in view: collect only what you need, keep it accurate, set a retention period. Anonymise transcripts where the identity of the participant is not essential to the analysis.

Set the deletion date when you collect, not when you finish. Deleting transcripts on a fixed date is easier than deciding later whether you still need them.

Good practice here is not only legal housekeeping. It also improves response quality, because participants who understand how their words will be used tend to give fuller answers.

Step seven: analyse, then write a brief people will use

Analysis means reducing, not summarising. Group comments into themes, count how often each theme appears, and note who said it. A theme raised by one participant is a hypothesis, not a finding.

Present findings against the decision from step one. Lead with the answer, then the evidence, then the caveats. Keep the deck short enough to be read in one sitting.

Translate each finding into an action with an owner. If a finding has no owner, it has no route into the work. That is how research dies.

The output most teams actually need is a persona document, not a report. Our guide to professional services marketing buyer personas explains how to turn themes into personas that writers and business development teams can use.

Keep the interview guide, the recruitment log, the lawful basis note and the persona document. Reused across cycles, they build an evidence base that survives staff turnover, the constant risk in a partnership.

Where official data adds weight

The public sector has published decades of method guidance that transfers well to commercial research. The GOV.UK service manual on user research covers recruitment, consent, session structure and analysis in plain terms.

Use it when you need to defend your method to a sceptical partner. It is easier to justify a fifteen-interview programme when the approach mirrors a documented standard.

It also helps with ethics. The manual's emphasis on informed consent and on not over-claiming from small samples is exactly the discipline a professional services firm needs when the stakes are reputational as well as commercial.

The same logic applies to advertising claims. The ASA and CAP publish research into advertising effectiveness and regulation, which is worth reading if your findings will inform campaign messaging that must stay within the advertising codes.

If you build messaging on audience research, keep the underlying data. A claim you cannot evidence is a risk you have chosen to take.

Desk research sources worth bookmarking

Desk research is cheap and fast. It sharpens your hypotheses before you spend money on fieldwork.

Use official statistics for market size and cost context, regulator publications for compliance constraints, and trade body reports for sector sentiment. Always note the publication date and the base the figure is drawn from.

Check the base before you quote a figure to a partner. A survey of 200 firms in one city does not describe a national market, and someone in the room will notice.

For competitor positioning, read the last twelve months of a rival's insight pages, tender wins and recruitment ads. Recruitment ads reveal internal priorities faster than any brochure. Our guide to professional services marketing competitor research shows how to structure that review.

How segmentation platforms change what you can ask

Paid platforms now offer targeting by job function, seniority, industry and employer. LinkedIn's audience targeting for professional advertisers is a useful example, because it shows what attributes the platform will let you select.

That matters for research design. If a platform cannot target the segment you have defined, your segment is hard to reach at reasonable cost. Reachability is part of a good definition.

It matters for validation too. You can test whether a message lands with a defined group without a full survey, using small paid tests as a signal rather than proof.

Treat platform data as directional. It tells you who is reachable and what they respond to, not why. The why still needs interviews.

A glossary for brief-writing

Common questions

How many interviews are enough?

For a single client type, twelve to fifteen conversations usually surface the main themes, and you stop when new interviews repeat what you have already heard. If your buying group has four distinct roles, treat each as its own group and recruit separately.

Can we reuse client data for research without asking again?

Often yes, if you have a lawful basis such as legitimate interests and your privacy notice covers research. Check the purpose you originally collected the data for, and consult the ICO guidance before relying on it.

Should we buy a survey panel or use our own list?

Use your own list when you need depth about your clients. Buy a panel when you need market-level proportions and can accept less context. Report which you used, because the two answer different questions.

How often should we repeat the research?

Run a light cycle quarterly and a fuller study annually. Client needs move with budgets and regulation, so a finding from eighteen months ago may no longer hold. Our guide to professional services marketing trends and outlook covers what to watch between cycles.