DEV Community

Vitalii Ermolaev
Vitalii Ermolaev

Posted on

Research methods for a product manager

One of the most important skills of a product manager is the ability to do research. They help to find out about «what is bothering» the user, to test a hypothesis, or to make a decision to launch a new feature. In this article, I will tell you about the main types of research.

Types of Research

Only three types of research are known: qualitative, quantitative and mixed.
Quantitative research is a kind of numerical value that determines the behavior of people and their attitude to something. Qualitative research works with non-numerical values ​​that were obtained through interviews, for example. Mixed research include the first two types of research.

Each type of research has a subtype: relationship research and behavioral research. The first research allows you to find out what is important for the user and what he wants. The second studies make it possible to observe how he acts in practice.

Qualitative Research

Interview

Subtype: Relationship Research, Behavioral Research

Interviewing is one of the most versatile research methods. It's used to investigate user behavior, identify their «pains» and needs. Interviews can be conducted at all work stages on a product, from forming and testing a hypothesis to maintaining a working service and introducing new functions. The advantage of this method is that it allows you to deeply study user motivation, better understand the target audience and get unexpected insights from it.

Usability Tests 

Subtype: Behavioral Research

Usability tests are research in which product managers observe how the user uses the service and what features are most important to him. 

Also, a person is given a task that he must complete, for example, setting up ads for certain groups of users. This research method is good because it allows you to quickly immerse yourself in the product, and it is especially useful for those managers who have recently moved to a new company or service.

Usability tests can be moderated when you are present in the research, communicate with the respondent and observe him, as well as unmoderated for the experiment purity. Thus, you can test your products and those of competitors to understand what the pros and cons are of the current solutions. Another advantage of usability tests is that they are easy to conduct remotely, for example using TeamViewer, and observe what the user is doing.

Focus Groups

Subtype: Behavioral Research

Focus groups allow you to identify the main problems of the product. The in-depth interview method is used during focus groups, but the product manager does not communicate with one user, but with a group of several people. This research method allows you to reduce the interviewing cost (due to less time consumption) and identify key issues. The disadvantage of focus groups is that group dynamics can shift the bottom line, for example, if a leader appears who will push others to his opinion. Therefore, it's up to the researcher to direct the group discussions.

Mixed Research

Surveys

Subtype: Relational Research

Surveys are flexible user research tools. They can be implemented in a variety of contexts: by posting a simple open survey on a website, through a mailing list, or after testing the usability of the service.

Surveys can provide quantitative and qualitative data: ratings, percentage of choice of one of the answers to a multiple-choice question, and, in addition, answers to open-ended questions. 

Card Sorting

Subtype: Behavioral Research

Card sorting allows you to form the menu structure or directory by the hands of users. During the research, the names of the objects to be sorted are recorded on cards, and then the respondents sort them based on their ideas about the structure of objects and the relationships between them. The researchers then analyze the results and identify patterns. The already existing hierarchical structure is checked for «reverse» card sorting. Participants are given tasks and asked to find a suitable card, moving through the category tree from the top to the bottom. This test verifies the health of the structure, for example, the location of the sections in the app and the relationships between them.

This research can be carried out in person and with physical maps, or remotely, for example through the OptimalSort card sorting platform.

Concept Testing with MVP

Subtype: Relational Research

An MVP is the simplest working prototype of a service that tests demand before starting full-scale development. This approach insures the company from creating an unclaimed product and helps not to waste resources on design and development in vain. MVP allows you to collect information with minimal effort in order to finalize the product in accordance with the needs of the target audience or completely abandon an unsuccessful idea.

Analysis of User Sessions

Subtype: Relational Research

During the research, they analyze the accumulated information about user activity, actions that the user performs while working with the service, such as queries in a search engine or transitions between sections in the app. Using session analysis, you can find out the scenarios of user behavior and simplify the achievement of their goals, for example, by reducing the number of steps required to complete an order.

Eye Tracking Testing

Subtype: Behavioral Research

Special equipment is required for eye tracking research, which tracks the movement of the user's gaze along the interface. When 30 or more participants complete the same task on the screen, you can notice patterns and figure out which elements on the page grab the user's attention.

Eye tracking helps you determine which elements of content you need to focus on and which elements are best distracted from.

The biggest drawback of eye tracking research is the need to use expensive equipment that can't be used without special training.

Quantitative Research

A/B Testing

Subtype: Behavioral Research

A/B testing is an integral part of working on any service. You can use it to test a hypothesis about whether the selected product metric will change if you change something in the product, for example, whether a change in the design of the registration page will increase the number of users. The results of the test and control groups of users are compared: the first group is shown a new solution, and the control group is shown an unchanged product.

And it's important for you to test whether the change will be statistically significant to confirm that the observed difference between the test and control groups is indeed due to product innovation and not an accident.

Clustering High-Quality Requests

Subtype: Relational Research

Clustering quality requests is a way to translate verbal requests and user feedback into quality data. This method allows using machine learning and natural language processing models to isolate requests that are constantly repeated in feedback.

Web Analytics

Subtype: Behavioral Research

Web analytics allows you to identify and prioritize product issues. It will help you find out the number of people who came to your resource, the depth of view (the number of pages viewed by visitors at one time), the conversion in paying or registered users, and so on.

Latest comments (0)