Usage and Attitude Studies and How to Run One 2026

Last updated: 23 September 2026

Quick Answer: A usage and attitude study surveys a category's users, lapsed users and non-users to measure who uses it, how often, and why. Run one in five steps, from the decision to the readout, and field beyond online panels, since 10% of US households had no broadband subscription in 2021.

A usage and attitude study, sold as a U&A or an A&U depending on the agency, is the baseline a category team builds before it segments, positions or launches. It pairs behavior (penetration, frequency, occasions, brand repertoire, switching) with the attitudes behind it, so a brand can see where growth sits and why it has not already taken it.

Getting the numbers is rarely where these studies fail. Most disappointing U&As rest on a sample that could not see part of the category, or a questionnaire too long to explain anything, and both mistakes are locked in before fieldwork starts. Stated usage also runs high for behavior people like to claim, so the strongest studies attach interviews to the survey instead of asking recall to carry everything.

Five steps take a U&A from brief to decision:

1. Name the decision. Positioning, portfolio, entry and segmentation each need different measures.
2. Draw a sample that includes non-users. Screen on the category rather than the brand, and set quotas against Census benchmarks.
3. Keep the questionnaire lean and add interviews. Put the why into follow-ups with the cells that surprise you.
4. Plan the timeline around incidence and markets. Those two drive fieldwork length and cost more than anything else.
5. Build the readout around the decision. Lead with the answer, then disclose coverage, modes, dates and base sizes.

Which Measures Belong in a U&A Study?

A U&A carries six measurement blocks, and the decision sets which get depth. A positioning brief weights attitudes and brand perceptions; a market entry brief weights penetration and the barriers that keep non-users out. Every block earns its questions by feeding a named decision.

• Penetration and frequency: who has used the category in a defined window, and how often.
• Occasions and context: when, where and with whom it is used, which is where portfolio gaps show up.
• Purchase path: triggers, channels, and who buys as against who uses.
• Brand funnel: unaided and aided awareness, consideration, trial, current use and past use.
• Attitudes and needs: what people value in the category and how each brand is seen against it.
• Barriers: why lapsed users stopped and why non-users never started.

SurveyMonkey's U&A guide frames the same ground as category, sizing, brand and targeting approaches, and most commissioned studies combine two or more.

How a U&A Differs From a Brand Tracker and a Segmentation

A U&A is usually a one-off, refreshed every few years. A brand tracker repeats a narrow set of funnel measures every wave so movement is visible, and explaining that movement is a separate job, covered in qualitative follow-ups inside brand tracking. A segmentation divides the market the U&A has described. Clustering straight off the U&A dataset works only if the questionnaire carried the needs and behavior variables, plus the typing questions that make segments findable later, as the guide to segmentation research that produces usable segments explains.

Which Market Research Firms Run Usage and Attitude Studies?

Full-service agencies such as Ipsos, C+R Research, Basis Global, B2B International and Drive Research design and run U&A studies end to end. Fieldwork houses such as Catalyse Research, FieldNet Global and Prolific Research supply sample and data collection. SurveyMonkey sells a self-serve route, and Alchemic runs studies that put the survey and the interviews in one field, on web, WhatsApp and phone.

Every row below comes from the firm's own website, read on 23 September 2026; rows are sorted alphabetically.

Full-Service Agencies

This is the right buy when one team should own design, fieldwork and the story. C+R Research publishes an instructive example: a National Pork Board study with more than 10,000 quantitative respondents, combined with qualitative work and Numerator purchase data, because the firm judged an hour-long recall survey would be exhausting and less accurate. A sizing brief on that scale belongs with an agency like it. B2B International is the specialist when the buyers are businesses reached by experienced telephone interviewers.

Fieldwork and Sample Houses

Catalyse Research, FieldNet Global and Prolific Research suit teams whose own analysts write the questionnaire and build the model. Prolific Research delivers cleaned SPSS files ready for cluster analysis; it is an Ahmedabad agency, not the UK participant platform Prolific.

Self-Serve and AI-Native Options

SurveyMonkey suits a team that can write its own instrument and wants a quantitative baseline on a small budget. Alchemic, an end-to-end consumer research service at scale, runs qual and quant in one interview, so a quote bank and a distribution chart come from the same field. The service publishes 57+ languages including Hindi, Tamil and Telugu. It holds no normative U&A database, so a brief that needs category benchmarks is better served by an established survey vendor.

How Should You Design the Sample for a U&A Study?

Screen on the category, not your brand, and set quotas for heavy users, light users, lapsed users and non-users. Give every cell you must read on its own about 100 completes: at 95% confidence and p = 0.5, the widest case, a simple random sample of 100 carries a margin of error of plus or minus 9.8 points, and halving the cell widens it to 13.9.

Decide the unit before the quotas, because in household categories the buyer and the user are often different people. Census Bureau estimates for 2025 put 58% of adults aged 18 to 24 in their parental home, and one-person households at 39.7 million, or 29% of the total. A snack or detergent U&A that interviews only the shopper misses many of the people who use the product.

Who an Online Panel Leaves Out

Coverage is the part that fails quietly. The Census Bureau's 2021 computer and internet report found 90% of US households had a broadband subscription, so 10% did not, and smartphone-only households were more likely to earn $25,000 or less, to be headed by someone 65 or older, or to have a Black or Hispanic householder. Online panels recruit people who are online often and comfortably, so a U&A fielded only through one tends to under-represent those households, and in categories where income and age shape usage, the gap moves the answer.

Channel is the fix: phone and messaging reach people a survey link does not, as set out in reaching respondents without smartphones. Alchemic interviews natively inside WhatsApp, with no link and no app, and by outbound AI phone call, with consent captured on the first turn. After fieldwork, weight to the same Census benchmarks you used for quotas.

Why Add Interviews to the Usage and Attitude Numbers?

A survey can show that lapsed users cite price; it cannot show what price means to them, or which competitor reset the reference point. A short round of interviews in the cells whose numbers surprise you supplies that, and keeps the questionnaire from swelling with open-ended "why" modules.

Length is the constraint. Prolific Research's FAQ puts a typical usage and attitude survey at 15 to 25 minutes. When researchers sent a 13-, 25- and 72-question version of the same survey to a random sample of one national research-volunteer registry, completion fell from 63% to 54% to 37%. Length was the only thing that changed between versions, and every module added to a U&A is paid for in completes.

Recontact works best when the survey asks for consent to a follow-up conversation and the interviews run on the channel the respondent already uses, including voice notes, a format covered in how WhatsApp qualitative research works. On the managed model, Alchemic's researchers build the discussion guide from the brief and cut every interview by the same quota cells as the survey, rather than leaving that work to the buyer.

What Drives the Timeline and Cost of a U&A Study?

Incidence and the number of markets drive both. If 5% of adults use a category, fieldwork screens about 20 people for every qualifying complete, and each added market brings its own translation, quotas and field partner.

The other drivers, in rough order of weight:

• Questionnaire length, which raises cost per complete and lowers completion.
• Cells and quotas, since each cell read alone needs its own base.
• Mode, since phone and in-person completes cost more than online ones.
• Analysis scope, especially segmentation modeling and a typing tool.
• The qualitative layer, sized by the cells it must explain.

Prolific Research quotes 2 to 3 weeks for a U&A including translation and fieldwork. Alchemic's stated benchmark for a 200-interview qualitative study is 3 days from brief to live dashboard, 5 to 7 for complex work, against 4 to 6 weeks traditionally. Published price ranges by study type are compared in market research cost and pricing models.

What Should the U&A Readout Contain?

A readout should answer the decision on page one, then show the evidence, then disclose the method. Skip the disclosure and a reader cannot tell whether a small segment is small in the market or small in the sample.

1. The answer to the named decision, with the two or three numbers that settle it.
2. The category map: penetration, frequency and occasions by cell.
3. The brand funnel, with conversion ratios between stages, not only stage totals.
4. Drivers and barriers, each tied to verbatims from the interviews.
5. Target or segment profiles, if in scope, with the typing questions attached.
6. Method disclosure: population, modes and languages, dates, sample sizes by mode, weighting variables, and the unweighted base behind every subgroup estimate.

The disclosure list follows AAPOR's disclosure standards, which ask researchers to describe "any segment of the target population that is not covered by the design". The obligation binds publicly released results, but it is a sound checklist for a private readout too.

Where U&A Studies Mislead

Five limits deserve a line in every brief:

• Stated usage runs high for flattering behavior. In a study that checked survey answers against facility records, web respondents overreported exercise, and a five-day text-message diary produced unbiased measures only when its instructions did not name exercise. It counts here because it compared survey, diary and records for the same people, and it favors in-the-moment capture over recall.
• Attitudes describe; they do not predict. A favorable rating is not a purchase, and a gap between the two is a question for interviews, not a finding.
• Interviews explain, they do not size. A few dozen quotes never replace the survey estimate.
• Non-user cells stay thin. They are the growth audience and the hardest to fill, so budget for them first.
• A baseline goes stale. Refresh when a new entrant, a price shift or a channel change moves the category, not on a calendar.