Usage and attitude study: turn category habits into a growth decision

A category can contain enthusiastic buyers who use it rarely, frequent users who choose a competitor and nonusers who have a practical reason to stay away. An average appeal score hides those differences. Before deciding to recruit new buyers or encourage another occasion, find out which situation the business is actually facing.

A usage and attitude study, often called a U&A or A&U study, connects category participation, buying and use patterns with the attitudes around them. Its useful output is a category baseline that helps a team choose a specific growth question to investigate. It should tell you whose behavior is represented, what counts as use and which explanation still needs a test.

Consider Plover, a fictional frozen-dumpling brand. Its team is deciding whether to investigate more weekday meal occasions among existing category users or barriers keeping potential buyers out of the category. Plover and the possible patterns below are illustrations. No study has been run and no measured findings are reported.

Choose the growth question before building the questionnaire

Plover sells a bag of 20 frozen vegetable dumplings through supermarkets. The team has two possible explanations for limited growth. Perhaps existing category users already like the product but think of dumplings mainly as an occasional shared snack. Alternatively, people who could use them for weekday meals may not enter the category at all.

Those explanations call for different work. The first might lead to a useful meal idea or preparation support. The second could involve product understanding, availability, freezer space or a mismatch with what people want to eat. A U&A should help decide which explanation deserves further investigation.

SurveyMonkey’s U&A guide describes the method as connecting who buys and uses a category, where and how often, with opinions and consumption habits. For Plover, that means studying the category beyond its own customer list. Surveying only Plover buyers would miss people choosing another brand and people not using frozen dumplings.

The commissioning decision is therefore specific: establish the current category baseline, then choose whether the next research should address an unused occasion among category users or a barrier to category entry. The baseline does not itself prove that either intervention will grow sales.

Define the category and the period people must recall

Decide what counts before asking how often it happens. In this illustration, category use means eating frozen filled dumplings prepared at home. Restaurant dumplings and other frozen meals are alternatives, rather than category-use occasions. Explain the boundary in language people understand and check it in a pilot.

Keep purchase and consumption separate. One person may buy the bag, another may prepare it and several people may eat it. A purchase is not an eating occasion, and a bag is not a person. Record the relevant role rather than asking one household member to guess everyone else’s behavior.

Use a stated reference period for frequency, with actual start and end dates in the fielded questionnaire. Ask about the most recent relevant occasion separately. The period needs to suit the category and the decision; four weeks is an illustrative choice here, not a universal rule.

The UK Government Analysis Function’s questionnaire design guidance recommends specific time boundaries and attention to recall difficulty. Where appropriate and consented, a participant can check a receipt or another record. An unsupported recollection remains reported behavior, even when entered through a digital survey.

For a category-use estimate, define the eligible population before filtering to current users. If the target is adults who share responsibility for household food shopping, ask enough to identify current category users, lapsed users and nonusers within that population. An oversample of occasional users may support a diagnostic, but it needs to remain identifiable when estimating the category baseline. Our research sampling guide covers the audience and inference decisions behind that choice.

Keep people, households and occasions in their own bases

Plover needs more than a count of bags sold. It needs to know how category participation and frequency relate to the opportunity under consideration.

A survey of one recent occasion per user produces a distribution of those users’ last occasions. It does not automatically measure the share of all category occasions: a daily user and a once-a-month user each contribute one record. Use an appropriate frequency or diary design when the decision requires the total occasion mix.

Keep that distinction in the report. “Few people use the category” and “users use it infrequently” are different statements. A team cannot choose between category recruitment and occasion development from a frequency average that quietly removed everyone reporting zero use.

Record the repertoire before asking for a favorite

People may alternate between frozen dumplings, other frozen meals, a meal cooked from ingredients and takeaway. Within the category, they may buy more than one brand. Ask about the alternatives used in the defined period rather than forcing a single favorite to stand in for their repertoire.

A person who reports using Plover and another dumpling brand belongs in both brand-use groups when the question permits multiple answers. Those percentages need not add to 100%. That is different from allocating every recorded purchase or occasion to one option. Preserve the question and denominator so the reader can tell which measure is being reported.

Separate what was available from what was chosen where the evidence permits it. A shopper may prefer another product, or Plover may simply have been absent from the store. Ask about the relevant encounter and retain the source of the account. A survey explanation alone does not establish store stock or the competitor’s distribution.

This category repertoire is an input to later decisions. It is not yet a segment structure, a customer persona or a recommendation to target every person who named an alternative.

Ask attitudes about the same use situation

An answer to “Dumplings are convenient” can refer to storage, preparation, serving or shopping. For the weekday-meal question, investigate the part that could change the next action.

One illustrative statement is: “Frozen dumplings fit the time I have to prepare a weekday evening meal.” Keep a separate question for confidence in making them into a complete meal. Combining the two would make disagreement hard to interpret. Use response options suited to the question, and keep lack of experience or inability to judge distinct from a neutral opinion.

Ask about category habits before introducing Plover’s proposed meal idea. Once someone has seen a persuasive explanation, a later answer concerns that informed encounter. Pew Research Center’s guidance on question order explains how earlier questions can supply context for later answers. Keep the original category baseline separate from responses after the proposed intervention.

Compare attitudes within the defined usage groups and read the associated accounts. Positive views with little use might point to an absent occasion, a practical constraint or an inaccurate explanation. The association does not establish which belief caused the behavior. Our consumer behavior guide explains how reported, observed and transactional evidence answer different parts of that question.

Let the pattern change the next study

These are conditional interpretations, not Plover findings. Their value is that they lead to different research and product decisions.

The third pattern is a useful countercase. It prevents the team from turning every U&A answer into a communication problem. If the target household cannot store the product, an occasion-led campaign may reach the intended person without making the offer usable.

Likewise, a large category-use group is not automatically the best growth opportunity. Consider the size and uncertainty of the measured group, the unmet situation, the feasibility of the intervention and the evidence needed to evaluate it. Preserve unavailable information rather than assigning an invented opportunity score.

Carry one opportunity into a separate test

Suppose the real baseline and follow-up accounts eventually support investigating weekday use among current category users. Plover could compare its complete existing presentation with a feasible meal-serving presentation of the same bag. Keep product, quantity, price, availability and exposure comparable when the question concerns the presentation.

That comparison would ask whether the proposed explanation changes understanding and consideration in the specified situation. It would not show that people can prepare the meal successfully or that they will buy more often. Test the actual preparation with intended users, then use an appropriate behavioral or retail design when the decision depends on purchases or repeated use.

If the proposed solution changes the product or adds another item, describe that as a different complete offer and evaluate its feasibility. Our concept testing guide covers the development decision after the opportunity has been selected.

The U&A earns its place by establishing which category question should reach that test. It should not become a concept score collected before the team knows which population or occasion it needs to understand.

Use the baseline as context for a Sapien study

Sapien builds simulated audiences grounded in purchase, behavioral and research data. Your customer data and past research can add context. We can use an agreed category question to compare alternatives, investigate modeled responses across audiences and explore the reasons behind them.

For Plover’s decision, an existing human U&A could inform the audience and the usage situations represented in a subsequent comparison. Keep the source, period, category definition and important gaps visible. A simulated account of eating dumplings is not a new human usage record. Generated frequencies cannot establish current category participation, the actual occasion mix or a change in real household behavior.

Our synthetic vs human respondents guide explains how to choose the evidence source as the decision changes. Use human and behavioral evidence for the category baseline and real-use outcomes; use modeled comparisons with appropriate grounding and validation to explore the next alternatives.

Bring the category, the growth decision and the customer or research evidence you already hold. Book a demo to see how Sapien can compare the next offers or messages, investigate audience differences and help your team decide what to test.