Concept and product testing: ideas, packaging and names

Questions this report answers

• Which concept has the strongest appeal and purchase intent among category buyers?
• Does the idea solve a problem buyers recognise, and do they believe the benefit?
• Which pack design is found fastest on shelf and signals the right flavour or variant?
• Which name is easiest to recall, fits the positioning and carries no unwanted meanings?
• What exactly puts people off, and can it be fixed before launch?
• Does the product deliver on the concept's promise once people have tried it?

When you need it

Concept testing earns its place at the points where development money is about to be committed. The most common are narrowing an innovation pipeline from several ideas to one or two; signing off a pack redesign, where the risk is that loyal buyers no longer recognise the product; choosing a name for a new brand or range; and checking that a cost-saving reformulation is not noticed, or not disliked, by regular users.

Most concept, pack and name tests now run as online surveys. Eurostat data for 2025 show that 89.5% of individuals in the EU-27 used the internet every day, with 82.7% in Bulgaria and 98.4% in the Netherlands. Coverage is therefore good for most consumer targets, although older or rural audiences in some markets may still need a share of face-to-face or telephone interviews.

A quantitative test is the wrong first step when the idea is still a sketch. In that case a round of focus groups or interviews will shape the proposition faster. Testing also adds little for small, low-risk line extensions, where a limited live launch teaches more than a survey.

Sample table of contents

1. Executive summary
   a verdict on each concept against the action standards
2. Objectives and action standards
   the pass marks agreed before fieldwork
3. Methodology and sample
   test design, cells, quotas and fieldwork dates
4. Concept scorecard
   appeal, relevance, credibility, distinctiveness and purchase intent
5. Diagnostics
   likes, dislikes and confusing elements from open answers
6. Pack and shelf results
   findability, standout and communication on a virtual shelf
7. Name results
   recall, fit and associations in each market
8. Product experience
   ratings after use in an in-home test
9. Source of volume
   which products the new item would replace
10. Recommendations
    which concept to take forward and what to change first

Methodology and sources

Monadic or sequential monadic?

In a monadic concept test each respondent sees a single concept. It gives the cleanest read, has no order effects and mirrors how a shopper meets a new product, but every concept needs its own cell, so the sample grows with the number of options. In a sequential monadic design each respondent rates two or three concepts in rotated order. It needs fewer interviews and allows a direct preference question, yet later concepts are judged against earlier ones, so we use it for screening rather than for the final go/no-go.

Action standards

Before fieldwork we agree what counts as a pass: for example, top-two-box purchase intent above the current product's score or a benchmark concept tested under the same conditions, with the difference significant at 95% confidence.

Stimulus, sample and checks

Concept boards follow one template, with headline, consumer insight, benefit, reason to believe and image, so that the idea is tested rather than the copywriting. A packaging test places each design on a virtual shelf among competitors. Respondents are screened category buyers; we remove speeders, add attention checks and code every open answer.

What you receive

• Report in PDF with a scorecard for each concept against the action standards
• Topline results within a few days of fieldwork closing
• Excel tables by segment, country and current brand used
• Coded open answers with representative verbatims
• Click heatmaps showing what people noticed on the board or pack
• Debrief presentation with a go, fix or stop recommendation

How the project runs

1. Briefing and action standards — 2–3 days
2. Stimulus preparation — 3–7 days to bring concept boards, pack renders or name lists into one format
3. Questionnaire, scripting and pilot — 3–4 days
4. Fieldwork — 5–10 days online, or 2–3 weeks when an in-home product test is added
5. Analysis and coding — about a week
6. Report and debrief — 2–3 days

Common mistakes and how to avoid them

• Testing a polished board next to a rough sketch. The better-produced stimulus wins, so bring every concept to the same standard.
• Setting the pass mark after seeing the data. Without action standards agreed upfront, any result can be argued either way.
• Relying on purchase intent alone. Stated intent overstates trial; read it alongside credibility and the diagnostics.
• Making the final decision from a sequential test. Order effects distort absolute scores; confirm the winner in monadic cells.
• Showing packs on a white background. Standout only means something on a realistic shelf among real competitors.

Who it suits

Updated: 5 Nov 2026