# Remote research Compare moderated and unmoderated research and learn when each format fits. All the research we discuss below takes place online. It can be moderated or unmoderated, and also qualitative or quantitative. Each format has its own purpose and its own place in the life of a product. ## What moderated research is In moderated research, the researcher communicates with the respondent directly — through video calls or messengers. These are always qualitative methods that help you understand users’ motivations, context, and barriers more deeply. The main formats include: - Moderated usability tests — check how users complete tasks in a prototype or product. - In-depth interviews — explore experience and perception in detail. - Focus groups — let you discuss a product or idea with a small group. - Diary studies — capture people’s experience over several days or weeks. - Observations — show real behavior in a natural environment. ## More about moderated usability tests In moderated usability tests, the researcher meets with each user and runs the session using a guide. This format works when you need not only to measure scenario completion, but also to understand motivation, expectations, and barriers. The advantage of moderated tests is that they let you work even with “rough” or imperfect prototypes. There are two approaches: - Structured testing — you have clear tasks and a fixed order for completing them. This format is useful for comparing participants and checking all hypotheses, but it requires knowing the user tasks in advance and leaves little room for unexpected insights. - Unstructured testing — the guide is approximate, and the researcher follows the respondent’s actions and tasks. This gives flexibility and can uncover unexpected insights, but the results are harder to compare with each other and there is a risk of moving away from the main research goal. Moderated research requires more time and resources than unmoderated research, but it provides a depth that is impossible to get without a live dialogue. ## What unmoderated research is Here, respondents complete tasks on their own: they go through a scenario, answer questions, and compare interfaces. This format can be both qualitative and quantitative — it all depends on the goal. A moderator is not needed during the study. They only need to prepare the study at the start — formulate the tasks correctly so the team gets the data it needs. [Methods and task design](https://wynde.evgeniichekov.com/en/guide/metody-kak-vybrat-i-zapustit#chek-list-dlya-proverki-zadaniy-i-voprosov-v-nemoderiruemyh-issledovaniyah) ## A new format — AI-moderated Here, the respondent also completes tasks on their own, but AI takes on the role of moderator. For example, Pathway AI can ask follow-up questions, run a mini-interview, and collect detailed answers. As a result, the researcher spends less time, and the insights become deeper. ## When unmoderated research is a good fit - Check whether the interface is clear: buttons, icons, and copy. - Compare several options, such as banners, screens, or messages, by attractiveness, clarity, or other metrics and, for example, select variants for an A/B test. - Understand how easy it is to find an element: account login, the “Buy” button. - Test a simple linear scenario: registration or checkout in e-commerce. - Check a prototype before release: find problems and optimize steps. - Validate usability-test findings quantitatively, for example by surveying more people to confirm a hypothesis about why a button label is confusing. - Validate a hypothesis quantitatively: survey more people after a usability test. - Run a “check” of a website or app: find bottlenecks and conversion barriers. - Check finished copy: marketing copy, value proposition, email. A good fit when: - you need to collect many responses quickly, - you need a diverse audience on a limited budget, - you need to observe behavior in a natural environment without moderator influence. ## When it is better to choose a moderated format - To understand needs, motives, and barriers, and to explore previous experience. In a survey you can ask about experience, or before an unmoderated test you can ask people to describe it, but without a live dialogue the result will be superficial. - To understand why an interface error occurs. An unmoderated test will show the result, but will not deeply explain the cause. An interview or moderated usability test lets you hear expectations and the context behind actions. - Long and complex scenarios are easier to test in a moderated format, especially when there are different branches and you need to understand the reasons behind actions. But if the scenario is linear or split into parts, respondents will not get confused, or the goal is only to measure completion, an unmoderated format can work too. - To work with an audience that is not used to online tools. If the product audience is not very “digital,” an unmoderated format can scare people away and make the results unnatural. - To discuss sensitive topics. In personal or emotionally difficult questions, people are more comfortable talking to a moderator. For example, in mental-health apps, users may share personal experiences, and without live contact with the researcher it may be harder for them to feel supported and safe. In general, unmoderated research is not suitable for every case where you need to talk to a user; for that, you need moderated research. ## Check yourself Choose the correct statement about unmoderated research. A. It can be used only for qualitative methods. B. It always requires a moderator to be present. C. It can be both qualitative and quantitative. D. It can be run only on finished products. Show the correct answer. Choose the incorrect statement. A. Unmoderated research is convenient on a limited budget. B. Moderated research is better for discussing sensitive topics. C. In long, branching scenarios, it is more convenient to use an unmoderated format. D. AI-moderated research can ask follow-up questions automatically. Show the correct answer. Match the task with the suitable method. Check whether users understand how to complete registration. Find out why people stop filling their cart. Discuss a new tariff with active customers. Options: A. Unmoderated test. B. Moderated usability test. C. Focus group. Show the correct answer.