Buyer persona research that finds what your buyers actually search

In short: A useful buyer persona connects buying decisions to real search behaviour. Start with customer conversations, identify the problems and questions behind a purchase, then validate those questions with search data and live results. Use the evidence to plan content that helps buyers make decisions, not just attracts visits.

What is a buyer persona in practical terms?

A buyer persona is an evidence-based description of a group of people who share relevant needs, responsibilities and buying behaviours.

For B2B content, its purpose is to explain what someone needs to understand before they can recommend, approve or buy your product.

A useful persona should tell you:

• What triggers the search for a solution.
• What problem the buyer needs to solve.
• What language they use to describe it.
• What alternatives they consider.
• What evidence they need to move forward.
• Who else influences the decision.

A fictional name and stock photograph will not answer those questions. Neither will a job title on its own. People with similar titles can have different priorities depending on their organisation, existing systems and immediate pressures.

Why do buyer personas often produce generic content?

Generic content usually starts with generic research.

A profile that says “time-poor leader who wants efficiency” could describe almost anyone. It gives a writer little direction beyond repeating familiar advice.

Search-led content needs a more specific chain:

Buying situation → practical problem → buyer question → search language → useful answer.

For example, a business considering workflow software might be struggling with handovers between teams. Its buyers may investigate approval delays, ownership rules, audit trails or integration requirements before searching for a product category.

Those are possible research directions, not proven keywords. The next job is to find out which questions buyers actually ask and how they search for answers.

This distinction matters when using AI. A plausible question generated by a model is a hypothesis. It becomes useful evidence only when checked against customer language, search data or other observable behaviour.

How do you research what buyers actually need?

Start with information your business already holds. You do not need an elaborate research programme before making progress.

Look for the moment that started the buying process

Review sales call notes, discovery recordings, support tickets, onboarding conversations and lost-deal feedback.

Ask customers and colleagues:

• What changed that made this problem worth solving?
• What had the buyer tried before speaking to us?
• What did they need to understand before agreeing to a meeting?
• Which concerns delayed the decision?
• What nearly caused them to choose another approach?

Capture exact phrases where possible. “We need better visibility” is less useful than an explanation of what someone cannot see and what happens as a result.

Handle recordings and customer information according to your organisation’s consent and privacy requirements.

Separate the roles involved

The person researching a solution may not control the budget. A technical reviewer may ask different questions from the daily user or senior approver.

Keep those perspectives distinct. Otherwise, your content may answer a user’s practical question while missing the security, implementation or commercial concerns that block a purchase.

Record evidence alongside assumptions

Maintain a simple research document containing the observed problem, supporting source, buyer role and unanswered questions.

Label assumptions clearly. Look for repeated patterns, but do not treat your loudest customer as representative of the whole market.

How do you turn buyer questions into search keywords?

Translate the research into groups of related searches without stripping away the original meaning.

Begin with a buyer’s own phrase, then explore variations involving the problem, task, solution category and buying concern.

For a workflow product, that might include:

• Problem language: “why do approvals get delayed”
• Task language: “how to track approval status”
• Category language: “approval workflow software”
• Evaluation language: “workflow software audit trail”

These are illustrative seed phrases, not validated search opportunities.

Check them using a keyword research tool with location settings relevant to your market. If you sell primarily in the UK, use UK search data rather than assuming global demand reflects your audience.

Then inspect the live search results. Look at whether Google returns guides, product pages, comparison articles, forums or something else. This helps reveal the likely intent behind a query.

Search volume is a useful signal, not a complete measure of value. Specialist B2B questions can have limited recorded demand and still matter to an active buying decision.

Content Foundry offers one way to connect this work: persona research, real UK search volumes and topic clustering can inform briefs, while its Company Brain supplies company knowledge and voice. Your team still needs to review the evidence and commercial relevance.

How do you choose topics that can support pipeline?

Prioritise questions where buyer need, business relevance and your ability to provide a credible answer overlap.

Use a practical set of checks:

Buying relevance: Does the question relate to a problem your product or service genuinely solves?

Decision value: Will answering it help someone assess options, reduce uncertainty or secure internal agreement?

Evidence available: Can you add experience, examples, product detail or subject expertise beyond what already ranks?

Intent fit: Is an article the right format, or would a product page, comparison or implementation guide serve the search better?

Next step: Is there a sensible action after reading, such as exploring an integration or reviewing an implementation approach?

Group closely related questions into topic clusters. Give each article a clear purpose so neighbouring pages do not repeat the same answer.

Avoid prioritising solely by search volume. Broad topics can attract people who will never buy. Equally, not every article needs an immediate sales call to action. Some should help buyers understand the problem well enough to take the next step.

How should buyer research shape an article brief?

A useful brief should make the reader’s situation unmistakable.

Include the buyer role, triggering problem, primary question, search intent, relevant objections and evidence needed. Specify what the article should help the reader decide or do.

Add the language buyers use, but do not force every variation into the copy. Review competing pages to understand expectations and gaps, not to reproduce their headings.

For answer-engine optimisation, make answers easy to find and interpret:

• Define unfamiliar terms plainly.
• Answer the main question near the beginning.
• Use descriptive headings.
• Explain processes in clear steps.
• Support claims with appropriate sources.
• Link to related pages where they help the reader.

These practices improve clarity and can support machine interpretation. They do not guarantee inclusion in AI Overviews or answers from ChatGPT. Useful content still needs accurate information, credible evidence and a reason to trust your organisation.

How do you know whether the research is working?

Measure progress beyond traffic.

Use search performance data to see whether pages appear for relevant queries. Check whether visitors continue to useful product, comparison or resource pages.

Ask sales whether prospects mention the content and whether it helps answer recurring objections. Where your tracking allows, examine content engagement alongside enquiries, qualified opportunities and pipeline.

Be cautious about attribution. A buyer may read an article, return through another channel and speak to sales later.

Use those signals to revise the persona, improve existing articles and identify missing questions. Research should evolve with buyer behaviour.

Related reading

• B2B content marketing strategy: from search demand to pipeline
• Content marketing for SaaS: what to publish when budget is tight

Frequently asked questions

What should a buyer persona include?

Include responsibilities, buying triggers, problems, questions, objections, decision criteria and research behaviour. Link important conclusions to evidence and label assumptions.

Can AI create a reliable buyer persona?

AI can organise evidence and suggest questions. Without customer research and validation, it can produce a convincing profile that is wrong.

Should every persona have separate content?

Only when its needs or search intent differ meaningfully. Shared questions can use shared content, with separate pages for genuinely distinct concerns.

How often should buyer personas be updated?

Review them when products, markets or buying conditions change, and whenever customer feedback or search performance challenges your assumptions.

Try Content Foundry free for 3 days, no card needed.