How to define a commerce reviews target audience

What to take away

A commerce reviews target audience is the group of people whose review-reading and review-writing behaviour you want to understand and influence.

Define the audience by the job they are doing, not by age band alone.
Split review readers from review writers, because each needs different evidence.
Size the group with published consumer data before paying for recruitment.
Give the definition a named owner and a review date.

Start with the job

Age, income and region are easy to pull from a dashboard. They rarely explain why somebody reads six reviews and leaves without buying.

Write the job in one sentence. For example: a person replacing a worn-out kitchen appliance who wants a model that survives a busy household. That sentence tells you which reviews matter.

The commerce reviews audience research guide for England sets out the wider research sequence, so read it before you lock a definition you will use for months.

Separate readers from writers

Most teams treat the audience as one crowd. It is at least two.

Readers vs writers

Review readers

Numbers: Outnumber writers
Wants: Volume, recency, durability
Trigger: Specifics about fit
Risk: Recruit wrong people

Review writers

Numbers: Wide margin fewer
Wants: A reason to bother
Trigger: Prompt after delivery
Risk: Recruit wrong people

Review readers outnumber review writers by a wide margin. Readers want volume, recency and specifics about durability or fit. Writers want a reason to bother, such as a prompt after delivery.

Blur the two and you recruit the wrong people. Name both groups separately in your notes.

Size the group first

Check what is already known about consumer behaviour before you commission anything. Which? publishes policy and insight work that helps you frame market size and shopping habits without buying a bespoke study. Its consumer data research is a sensible starting point.

Consumer protection rules are largely UK-wide, while some trading standards enforcement sits with local authorities. Say which nation your evidence covers.

Check what regulators know

The Advertising Standards Authority and the Committee of Advertising Practice publish research on how consumers understand online reviews and claims. That work can save you from asking questions that already have answers. See the ASA and CAP research library.

Use it to shape your questionnaire. If regulators have already tested how people read star ratings, you do not need to repeat that.

Map where reviews are read

Reviews reach people through product pages, marketplace listings, search results and map listings. If your audience finds you locally, platform documentation matters. The Places API legacy overview explains how place reviews are accessed.

Record the source and date of every figure you use in the definition.

Recruit a small panel

Eight to twelve people is enough for qualitative work. Recruit from recent buyers, lapsed buyers and people who read reviews but never write them.

Ask each person to talk through the last review they read in full. Listen for what made them trust or dismiss it. Record only with clear consent.

Turn raw notes into structured groups using the method in our guide to commerce reviews buyer personas in England.

Keep the definition current

Audiences move. A definition written for one product line drifts once you launch another. Put a review date in the document and name its owner.

Twice a year, check whether the job still matches what customers ask. If it does not, rewrite it rather than adding a segment.

Checklist

One sentence describing the job to be done.
Separate notes for readers and writers.
At least one published source cited with its date.
Regulator research checked before writing new questions.
Panel of eight to twelve people, recruited from three buyer states.
Consent recorded for any session notes or recordings.
Named owner and a review date within six months.

Common questions

How many people do I need for audience research?

For qualitative work, eight to twelve is usually enough to hear repeated themes. To quote a percentage you need a properly sampled survey.

Can analytics replace talking to people?

Analytics shows what happened, not why. Use it to size the audience, then use conversations to explain the behaviour.

Do I need separate definitions for the nations?

Not always. Consumer protection is largely consistent across the UK, but enforcement differs. State the coverage of your evidence so nobody assumes it applies everywhere.

How often should the definition be reviewed?

At least twice a year, and immediately after a product launch or a change in where reviews appear.