The Product Discovery Process – how to discover users’ needs

proces product discovery

You come up with an innovative idea, you write down your assumptions in a business plan… and then what? How to check if the idea that has emerged in your head is something your customers really need? What does the Product Discovery process have to do with it?

What is Product Discovery?

Product Discovery is one of the first phases in product development. The objective is to examine the assumptions concerning potential customers’ problems, to collect information validating the idea (positively or negatively), and to determine further design steps.

The main activities in this phase include research on new products. In the case of seeking optimization or feedback on the product (e.g. in the beta testing phase), you enter the next stages of the product life cycle, which means the next stage of research as well. It is possible to adopt an approach where the Product Discovery process does not end. After the definition phase, parallel to the design activities, the Continuous Product Design phase should follow, i.e. further discovery of needs.

The best and the easiest way to illustrate this is the Double Diamond design process model. The Product Discovery phase is represented by the first diamond. This is the stage of gathering as much information as possible about the target group (discovery) followed by elaboration, validation of hypotheses, and conclusions (define).

product discovery double diamond

Double Diamond design process model

What is the outcome of the Product Discovery process?

As a result of correct research and experiments conducted in this phase, you should obtain the information on:

  • users and their environment – who they are, what they need, what problems they have, and what values they hold.
  • problem – relevance, scale, and cause; how it affects the user; how it has been solved so far, and how you can solve it with your product,
  • project context – what is the correlation between the users’ needs and other spheres of their life or market, what other activities and problems influence the examined area,
  • common vision of the product – establishing the vision that meets your business needs as well as the needs of a target group,
  • validation of the initial idea – acquiring information that will help to determine whether your initial vision of the solution is desired by anyone, and what activities should be continued or stopped.

When is it necessary to conduct such research?

Research should occur whenever there is not enough confirmed information about user needs. When you have too little data to move forward with a project, and need to base your decisions on the actual need of the target group rather than your beliefs. The scope of the project and, at the same time, the subject of the research is not narrowed – you can work on creating a new product for the existing market,  innovative solutions, or a new strategy for the organization.

However, all these examples have one common denominator – you need to discover something new, find out about real problems and needs, validate initial hypotheses and assumptions. Going forward without research could be risky – you might be relying on your own beliefs too much. It may turn out that the “problem” you have identified is a problem only you can see. The team’s work will just be a waste of time and money. You will not meet the real need of your target group.

How to get started in Product Discovery – establishing the goal of the study

Regardless of whether we talk about Product Discovery research or efforts to improve an existing product, it is necessary to determine the purpose of the research (main goal and detailed goals). The following questions may be helpful:

  • What motivates us in this project? What makes us want to do the research?
  • What do we want to achieve through this research?
  • What is the intended outcome of the study and product?
  • What do we want to do after completing the research, what steps should we take?
  • What is the probable cause of the problem we want to examine?
  • What do we already know about the problem or the given field?
  • What are our actual time capabilities and budget?

Defining the goal allows you to set a clear direction and decide what you want to achieve (what you need to focus on). It gives the whole team a better comprehension of the final goal and minimizes the risk of misunderstanding. Additionally, arrangements related to the budget, deadline, and final result allow for a better definition of the initial action plan and facilitate selection of appropriate methods – after all, the research is to serve your end.

Once the main purpose of the activity has been identified, the next step should be to clearly define the problem under study and the initial plan of action to address that problem. 

Appropriate research plan – the core of activities

The first stages of research, especially in agile methodologies, can seem quite chaotic. It’s like landing up in the desert – you have no map, no information, no sense of direction, just a destination, and a few assumptions on where the water source might be. It is necessary to take a holistic look at the (product) environment and then, according to the observed signs, navigate (deepen the research) in the most promising direction.

A research plan (based on, among other things, hypotheses concerning the business model) should be prepared to structure the process. As the research progresses and user input is gathered, it will be refined on an ongoing basis. The overall framework of the research defining the goal, time and scope, budget, team capabilities, and method will allow you to focus on the actions leading to the goal. When establishing such a plan, keep in mind you can change the method on the fly to adjust it to the information gathered. While the general process is set before the work starts, adjusting the methods to the actual needs and research findings helps to execute the actions that make the most sense for your product. This way you focus on the actual needs without wasting time on unnecessary work.

So what does the research process look like in practice?

Appropriate adjustment of the process to the goals and other assumptions goes hand in hand with the research plan. How does this process work?

There is a myriad of activities and methods that can be used in discovery activities and they need to be reviewed on an ongoing basis. The question “what is our goal and will this method help us achieve it” should be raised repeatedly during the research process. Therefore, defining a fixed and linear research process in this phase is very difficult – after all, you choose to do research to learn about the unknown.

However, there is a basic division into specific research loops (for good reason similar to the loops in the Design Thinking process), where the methodology is selected depending on the needs. Each of these loops should end with a decision –  you either take the findings into the next phase (prototyping, testing, and then implementation), or you deepen them in the next round of research. Importantly, both activities can be performed simultaneously – in many cases, it is possible to transfer some of the validated hypotheses to the subsequent stage of development (e.g., the general outline of the product), while deepening the information further. The number of iterations in the loop should be determined individually based on the research needs and team capabilities.

Research process loop


Each loop in the research process begins and ends with a decision. At the beginning of your activities, you need to make some decisions about the methods and actions that will bring you closer to your goal. Any tool that facilitates research planning (e.g. Experiment Sprint Canvas) may be useful at this stage. 


The next step is empathization. Especially at the initial stage of research, try your best to understand your target users and “walk in their shoes” for a while. At this stage, the information to be explored is not yet confirmed – you are operating on hypotheses that you need to validate.

All the tools that help you understand the user and the project, such as protopersonas, empathy maps, or experience maps, are invaluable. This is also a great moment for a kick-off workshop, brainstorming sessions, and initial exploration of the topic. Keep in mind that the mapped information is the hypotheses that are yet to be validated – multiple revisions of the canvases are normal.


You have established a plan and goal of action, you are properly acquainted with the project, you have conducted preliminary exploration, empathization and you have defined research hypotheses. It is time for the observation phase (these phases often overlap). In this step, consistent with the purpose of a particular research loop, you acquire information to validate a hypothesis. This ensures that you don’t stray in the wrong direction.

Types of research in the observation phase

You can use many different research methods in the observation phase. They are usually divided into two categories: quantitative research (answering the questions of what, when, how many) and qualitative research (answering the questions of how and why). There are a huge variety of methods in each of these categories. Many of them combine both quantitative and qualitative features, depending on how you conduct the study.

The most popular and useful methods are:

  • Exploratory research (also known as generative research) – that is, research where you explore a topic, discover as much information as possible, but do not test anything (of course, the information gathered may indirectly validate some hypotheses). The scope of these activities is wide and includes the phase of desk research, field or Internet activities (ethno- and netnography), observations, exploratory interviews, or other modifications of popular research (even focus groups). At this stage, all the tools that allow for a better understanding of the whole context, such as mind maps, user journey, diaries, protopersonas, and affinity mapping are very useful.
  • Stakeholder interviews – that is, longer (but not exploratory) conversations with all the people involved in the project and interacting with the product. This includes interviews with target users as well as all team members who have or will have an impact on the product (e.g. company employees). All methods and tools that allow for prioritizing questions (such as Experiment Sprint Canvas), checking the validity of the research (pilot tests) or focus groups can be useful.
  • Workshops – conducted within the research team and with all project stakeholders or the target group under study. Workshops allow to gather information, map assumptions, and draw conclusions – workshops expand the information pool, enable to quickly learn about other people’s perspectives, and develop new solutions or hypotheses.
  • Quantitative methods – that is, all activities that allow you to validate a hypothesis and obtain information from a larger audience. They do not provide direct information about the hidden motives or needs of the interviewees, but they allow to validate hypotheses with a larger sample of participants and identify specific user groups. It is particularly recommended when combined with qualitative research such as interviews – for double validation.


The final step before returning to decision-making is interpretation. The collected feedback and enormous amount of data will only be useful if properly interpreted, mapped, and converted into specific theses (or further hypotheses). You can use the same tools as in the empathization and observation phases – map the information that has already been validated and tested with users.

At this stage, all the tools for data clustering (e.g., affinity mapping), determining patterns among target groups (e.g., personas, antipersonas) and user paths (e.g., user journey, empathy paths, service blueprint), identifying user values (e.g., Golden Circle, UVP Canvas) or user needs (e.g., user stories) will come in handy.

Return to the decision

The final steps before returning to the decision point are to determine if the information gathered needs to be deepened or if it (or some part of it) should enter the product development phase.

As mentioned above, you don’t have to stop research to use some of the conclusions in the design phase. Quite the opposite! Creating the basis of a project and simultaneous exploration and validation of subsequent detailed assumptions may accelerate the work and the product release… to validate it directly on the market. Of course, the decision of whether to take this step should be made on a case-by-case basis.

At this stage, all the tools used to prioritize the assumptions and collected information (e.g. MoSCoW, Eisenhower matrix, or Theme Scoring) will be indispensable.

Completion of Product Discovery – what’s next?

While (according to some studies) the Product Discovery phase in the product life cycle ends with the start of design activities (the development phase in the Double Diamond model), this does not mean the research should end too. Moving into the Continuous Product Design phase allows for further discovery of needs.

Additionally, research related to testing initial concepts, ongoing monitoring of feedback on the newly released product (e.g. in the beta-testing phase), and attempts at optimization (e.g. extending functionality or improving performance) should be undertaken. All these steps require specific research activities to minimize the risk and to build on real needs rather than on guesswork.

The time commitment and cost should be aligned with the goal and capabilities. You can gather feedback in usability testing, A/B testing, create new advanced prototypes, experiment, and so on.

What if you don’t have the time or funds to conduct full research?

Talk to a user or potential customer, conduct a simple survey, analyze all the information you get on an ongoing basis – simply do your best to hear the user’s voice and needs. Even a small number of user opinions and insights will be more valuable than working on an idea on your own without feedback from your target audience.

Additional materials


  • Don Norman, Design of everyday things
  • Teresa Torres, Continuous Discovery Habits
  • Testing Business Ideas: A Field Guide for Rapid Experimentation
  • Markus Edgar Hormess, This Is Service Design Doing
  • Marc Stickdorn, Jakob Schneider, This is Service Design Thinking
  • Louise Downe, Good Services: How to Design Services that Work

O’ Reilly collection, especially:

  • Farkas David, Nunnally Brad, UX Research. Practical Techniques for Designing Better Products
  • Jaime Levy, UX Strategy
  • Jim Kalbach, Mapping Experiences.

Internet sources

Articles aren't enough?

Would you like to learn about the whole process and how we could carry it out in your organization?

Let's talk
Agnieszka Zygmunt
Agnieszka performs analyzes, conducts research, and experiments. She specializes in Product Discovery research and helps create solutions focused on the real user needs.

She takes a holistic view of the design process accounting for t all factors that may affect the final user experience. In her work, she combines her analytical skills with a high level of empathy and love of science.

Co-creator of the Proste Rozmowy podcast. Privately involved in non-profit projects supporting those who need it most.

We eat cookies!

You don't believe it? Check out our cookie recipe and visit our office. We serve them with delicious coffee!