Before you brief a supplier, get your promotional merchandise audience research right

What to take away

Why the brief usually comes too early

The pattern repeats across England's promotional products trade. A marketing manager needs 500 branded items for a conference in six weeks. They ask three suppliers for quotes, pick the middle one, and the order ships. Nobody asked what the recipients would do with the item afterwards.

That is not a procurement failure. It is a research failure that shows up later as stock in a cupboard. The fix is to treat audience research as the first deliverable of the campaign, not an optional extra once the artwork is approved.

There is a second cost to skipping it. Suppliers price against uncertainty. If the brief leaves out the audience, the volumes and the delivery date, the quote carries contingency you end up paying for.

Research also protects the budget. If you know the audience is 200 account managers rather than 2,000 event visitors, you buy fewer, better items. The saving is rarely the point, but it makes the argument easier when finance asks why you spent two weeks before spending anything at all.

There is a simple test for whether a study is real or decorative. Ask who will read the output and which decision it changes. If the answer is nobody and none, it was never research.

Start with what you already own

Desk research comes first because it is free and fast. Pull the last three years of order data, enquiry notes and returns. Look for who requested merchandise, what they asked for, and what recipients said afterwards.

Customer relationship records are the richest source. A sales team that logs every giveaway as a note on the account has already done half your fieldwork. If those notes do not exist, that absence is itself a finding worth reporting.

The cheapest useful source is your own order history. It shows which items were reordered and which were quietly dropped. A reorder is the strongest signal a merchandise programme produces.

Look at what came back, not only what went out. Returned items, unclaimed stock and anything left on a table at the end of an event all say something about fit and usefulness.

Set a simple rule for what counts as evidence. A dated sales note attached to a named account is evidence. A colleague's memory that everyone loved the mugs is not. Write the rule down so the same standard applies to every claim in the final report.

Government research helps with risk context on the product side. The Department for Business and Trade's product safety research collection sets out how risk assessments are informed for consumer goods, which matters when you choose items the public will handle.

Map the audience before you map the product

Split your audience into two groups: the people who approve the spend, and the people who receive the item. They want different things, and confusing them is the most common source of wasted budget.

Approvers care about brand fit, cost per unit and whether the item will be used at an event they are attending. Recipients care about whether the item is useful on the day and whether it embarrasses them on the train home.

Write both groups on a single page before you contact any supplier. One page is enough, and it stops the brief drifting towards whatever the approving manager happens to like.

Event audiences and office audiences behave differently. An item handed out at a stand competes with a bag full of other items, so it has seconds to earn its place in the bag.

Internal gifting applies a different test. A recognition item given in front of a team carries meaning that a conference freebie never will. The same product can succeed in one setting and fail in the other.

For B2B campaigns, the approver is often a marketing or HR lead. LinkedIn's audience targeting documentation for advertisers explains how professional attributes can be used to reach specific job functions. That is a reasonable proxy for sketching who sits in the buying group before you speak to them.

Design interviews that produce usable answers

Interviews beat surveys at the start because you do not yet know which questions matter. Book 30 minutes with eight to twelve people across approvers, recipients and one or two suppliers.

Recruit from your own pipeline. Recent buyers, lapsed buyers and people who declined a quote all have something useful to say. The last group is the one most teams skip.

Ask about the last purchase, not the next one. "Tell me about the last item you ordered" produces a story. "What merchandise do you like?" produces a preference with no context.

Watch what people do as well as what they say. Ask them to describe where the last item lives now, or what happened to the box it arrived in.

Record the language people use. If three approvers say "something people keep on their desk", that phrase belongs in your brief. Our guide to running promotional merchandise customer interviews covers how to structure the sessions and avoid leading questions.

Choose survey tools that fit English data rules

Once you have themes, a survey can size them. Keep it under ten questions and under five minutes. Response rates collapse after that, particularly with internal distribution lists where people answer quickly and without care.

Decide which decision the survey will inform before you write a single question. A survey with no decision attached collects opinions nobody uses later.

Keep questions closed wherever you can, so the data cleans itself with little editing. One open question at the end captures the language you can reuse in copy and on stand graphics.

Write down the sample you are actually reaching. An internal list of 300 people inside one company is not a market sample, and the report should say so plainly.

If you collect names, email addresses or employer details, you are processing personal data and the UK GDPR applies. That means a lawful basis, a retention period and a way for people to opt out. Our review of promotional merchandise survey tools in England compares platforms on exactly those points.

Pilot the survey with five colleagues before it goes out. Every ambiguous question you catch at that stage saves a week of cleaning inconsistent answers later.

Turn findings into buyer personas

A persona is a short, evidence-backed description of a buying group, not a fictional character with a stock photo. Include the trigger for buying, the constraints on budget, and the objections that stall a decision.

Give each persona a name you will recognise in meetings and a one-line summary at the top. The rest is supporting detail for anyone who wants to see the evidence behind it.

Include the objections and the timing. A persona that says budget is approved in March is more useful than one that says the team values quality.

Three personas are usually enough for a merchandise programme: one approver, one recipient group and one procurement or finance gatekeeper. More than that and nobody uses them.

Every line in a persona should trace back to an interview, a survey result or a record. If you cannot say where a claim came from, cut it. The method is set out in more detail in our guide to building promotional merchandise buyer personas in England.

Worked example: a 300-person conference

A software firm in Manchester is exhibiting at a 300-delegate conference. The marketing lead has a merchandise budget of £4,500, or £15 per delegate if everyone takes one item.

Desk research shows last year's event produced 180 usable leads and 90 conversations. Interviews with six sales colleagues suggest the item should survive the journey home in a bag.

Interviews with ten past delegates put the useful threshold at under £8 per unit for something they would keep.

The plan splits: 300 low-cost items at £6 each, £1,800, plus 60 higher-value items at £25 each for booked meetings, £1,500. Freight and artwork £700. The total is £4,000, leaving £500 for a follow-up mailer.

The split follows the finding that most delegates want something small and most meetings are booked by a minority of visitors. Spending evenly across 300 people would have put a £15 item in every bag and left nothing for the follow-up.

Those figures are an illustrative example, not a market benchmark. Substitute your own unit costs, freight quotes and event data before you take anything to finance.

Check pay, safety and data rules before you rely on findings

Research findings age. Pay bands quoted in a persona may be out of date within a year, so check them against the current rate. Acas publishes guidance on pay and deductions covering the National Minimum Wage and related rules for employees in England, Scotland and Wales.

Data rules apply to the research itself, not only to the campaign. If you hold interview notes with names attached, they sit under the same retention rules as any other customer record.

Product choices carry their own checks. Items aimed at children, food-contact goods and anything with a battery need evidence before they go on a stand. Our promotional merchandise rules and compliance guide covers the UK-wide position on safety, claims and data.

Workforce context also shifts. The CIPD knowledge hub publishes research on employment trends and reward practice. That is useful when you are justifying gifting to an HR audience whose priorities have moved.

Build a research rhythm, not a one-off project

A single study is a snapshot. Promotional merchandise audience research works best as a repeating cycle, with a short review each quarter and a fuller one once a year.

Quarterly, review order data and run five interviews. Annually, run the survey again and refresh the personas. This keeps the evidence current without turning research into a permanent overhead.

Assign one person to own the cycle. Research that belongs to everybody is completed by nobody once the campaign goes live.

Keep a single document with the current findings and the date each was gathered. Anyone briefing a supplier can then quote a source rather than an opinion.

Log what changed after each cycle. Over three years that log becomes the strongest evidence you have that the programme earns its budget.

Write down what you will stop doing as well as what you will start. If the data shows tote bags are not taken home, removing them is a result, not a failure.

Brief the supplier with evidence attached

When the brief finally goes out, it should carry the research with it: audience size, item role, acceptable unit cost and the reason each item was chosen. Suppliers respond better to a clear problem than to a vague request for ideas.

Include the questions you still cannot answer. A supplier who knows the gap can suggest how to test an item at small scale before you commit to a full run.

Ask suppliers what they see across similar campaigns. They hold useful pattern data, though it is commercial rather than scientific. Treat it as a prompt for your own questions rather than a finding.

The merchandise sits inside a wider campaign. If your team needs a refresher on structuring that campaign, the Chartered Institute of Marketing runs strategy and planning training that covers objective setting and measurement.

Compare against what competitors actually do

Competitor activity is easy to observe and easy to misread. A rival handing out branded bottles at a show tells you they chose bottles, not that bottles worked.

Record dates and event names alongside what you saw. A pattern from three shows over two years is stronger than one observation from last month.

Collect what you can see, then test the assumption with your own audience. Ask recipients whether they noticed the item and what they did with it afterwards.

A structured approach to this is set out in our guide to promotional merchandise competitor research, which covers what to record and how to separate observation from inference.

Common questions

How many interviews do I need before I can trust the findings?

Eight to twelve is usually enough to surface the main themes for a single campaign. Stop when two consecutive interviews produce nothing new, and be honest in the report about how small the sample was.

Can I use a survey on its own without interviews first?

You can, but you will not know which questions to ask. Surveys are good at sizing themes that interviews have already identified, and poor at discovering themes from scratch.

Do I need consent to survey delegates at an event?

If you collect personal data, yes, you need a lawful basis and a clear notice. A short privacy line on the form, plus a retention period, covers most event research in England.

How often should personas be updated?

Review them annually, or sooner if the audience, the product range or the pay rules change. A persona built on 2024 data is a liability by 2027.