Beer consumer research methods for marketing teams
Beer marketing teams commission research on brands, packs and liquids. Each question suits a different method. This guide covers the main ones, the sample sizes behind them and the extra care a beer tasting needs.
Match the method to the question
Most research mistakes start with the wrong method. If you want to know why people choose a beer, you need qualitative work. If you want to know how many people would choose it, you need a survey. If you want to know whether they can taste a recipe change, you need a sensory test. These answer different questions and none of them substitutes for the others.
Start every brief with the decision the research will inform. "Should we launch a 500 ml can in Karnataka?" leads to a different design from "What do young drinkers think of our brand?"
Qualitative research finds the language
Focus groups, in-depth interviews and accompanied shopping trips are qualitative methods. They use small numbers of people, often six to eight per group, to explore attitudes, occasions and language. They are good for finding out how drinkers talk about beer, what a brand means to them and why they switch.
They are not good for measuring. Eight people liking a label does not mean most shoppers will. Use qualitative work to build hypotheses and the words to test, then measure with a survey.
In beer, occasion matters a great deal. The same person may choose a light lager with dinner and a strong beer at a party. Ask about specific recent occasions, not beer in general. "Tell me about the last time you bought beer" gets a truer answer than "What do you look for in a beer?"
Surveys measure, if the sample is right
Quantitative surveys measure how common an attitude or behaviour is. They need enough respondents for the result to be stable: as a rough guide, around 400 per group you want to compare gives a margin of error of about plus or minus 5 points on a 50% measure.
For concepts and packs, two designs are common. In a monadic test, each respondent sees one option, which avoids comparison bias but needs a bigger sample. In a sequential test, each respondent sees several options in a rotated order, which is cheaper but lets the first option colour the rest. Choice based methods, such as conjoint analysis, ask people to choose between packs with different prices and features. They estimate how much each feature drives choice.
Beer research must screen respondents for legal drinking age in their state and for recent beer drinking. Our guide to premiumisation in beer shows the kind of question surveys can help answer.
Sensory tests answer liquid questions
When the question is about the beer itself, such as a cheaper malt, a new hop supplier or a recipe tweak, use sensory methods. The triangle test is the classic one. Each taster gets three samples, two the same and one different. They pick the odd one out. Guessing gets it right one time in three.
A worked example: with 24 tasters, the standard tables need at least 13 correct answers to conclude, at the usual 5% significance level, that people can tell the beers apart. Fewer than 13 does not prove the beers are identical. It means this test found no difference.
Preference tests come after difference tests. There is no point asking which beer people prefer if they cannot tell them apart.
Beer tastings need extra care
Any research that involves drinking needs a responsible design. Participants must be of legal drinking age, must agree to take part and must be free to stop. Keep sample sizes small, often 30 to 60 ml per beer. Limit the number of beers in a session. Provide water, plain crackers and spit cups. Make clear that spitting is normal, not rude. Offer a non-alcoholic option so people who do not drink can still take part in the packaging and concept parts of the session. Nobody should drive after a session where they drank, so plan transport.
Serve beer at a consistent temperature and in identical, coded glasses so the brand and pack do not bias the result. A beer served warmer than another will taste different for that reason alone. If complaint data prompted the research, our guide to handling beer consumer complaints explains how quality teams investigate.
Where to start
Take your next research brief and write one line at the top: the decision it will inform and who will make it. Then name the method that answers that question and the sample it needs. If the brief cannot fill those three lines, it is not ready to go to an agency.
Common questions
What is a triangle test in beer research?
Tasters get three samples, two identical and one different. They pick the odd one out. If enough tasters are correct, the beers are detectably different.
When should beer brands use focus groups?
To explore attitudes, occasions and language. They are good for building hypotheses but too small to measure how common a view is.
How should consumer beer tastings be run responsibly?
Legal age and consent, small samples, few beers, water and spit cups, coded glasses and a clear plan so nobody drives afterwards.
