Understanding Users’ Needs: the Jobs-to-be-Done (JTBD) Approach
One of the key success factors of a mobile app is how well it meets users’ needs and helps solve their problems. But how do you identify the ideal product for your target audience? One of the methods we use for learning more about our users is known as Jobs-to-be-Done.
My name is Anna Naumenko. I’m the Lead Market Analyst at BP Mobile (AIBY Group). In this article, I’ll share our experience of putting the Jobs-to-be-Done methodology into practice. Below you’ll find a few handy templates for working with this technique that you can download and use in your projects.
What is the JTBD approach?
Jobs-to-be-Done (JTBD) is a method of target audience analysis that helps understand what tasks the product should solve for the user.
Another method — the target audience portrait — is still often used. This method assumes that potential customers are presented as a set of characteristics (e.g., age, sex, job). But the user is not a set of characteristics; they’re a living person with a task (a job to be done) who are looking for the best way to solve it. And in practice, it often turns out that these tasks are not as obvious as they may initially seem.
For example, users install our HitFit app not for workout programs, but rather for the possibility to improve their health and get “the dream body.” People exercise to be able to feel and look better. Fitness is the way to accomplish this task, and the mobile app is one option to get a workout program.
Understanding the underlying motivation of users helps you create a product that can solve their real tasks. This product is more likely to be in demand than an app filled with features that the creators thought were useful for their audience.
The JTBD method has proven particularly useful when creating a new product or when an existing product is stagnant and the team needs ideas on how to improve it.
Steps of the Research
Let’s take a closer look at each of the research steps according to the JTBD approach. I’ll also share some useful tips and examples to help you conduct your own research.
1. Setting Goals
First of all, define the goals and formulate the main hypotheses the project team wants to test. You can also add goals during the research.
2. Defining the Target Audience of the Research
Evidently, to get valuable information for your project, you should interview people who already use your product or potentially want to use it. Interviews with users of competing products will also be a valuable source of information.
First, define the main parameters of the audience you want to interview in your research (e.g., gender, age, location).
Next, make a list of screening questions. They’ll help you ensure that a potential respondent, who meets the basic parameters of your target audience (such as gender, age, etc.), can potentially be interested in your product and thus provide valuable insights for the purposes of research and testing hypotheses. From our experience, 3 to 5 product-themed questions are enough.
For example, while conducting research for Plantum (formerly NatureID)–an app that helps you identify plants by photos,–we selected people using the main parameters (location and age) and asked them a few questions to see whether these potential respondents like plants. In particular, we asked them if they enjoy taking photos of plants, learning more about them, taking care of them, etc.
Useful tip: A respondent may try to trick you and pretend that the topic of your research is of interest to them. To weed out such users, design your screening questions so that the correct answers are not obvious.
Here is an example of a screening question for the Plantum (formerly NatureID) app:
Which mobile app functions do you use?
a. Custom keyboards/fonts
b. Identification of plants by photos
c. Second phone number
d. None of the above
Respondents will be included in the survey only if they select answer b alone.
3. Creating a Questionnaire
The questionnaire should include questions that will help achieve research goals, test existing hypotheses, and generate new ones. It’s also recommended to focus on the fact that the main purpose of the research is to identify the tasks–jobs–the user employs the product for, as well as to understand the context of these jobs.
Useful tip: To make the interviewing process more flexible, divide the questionnaire into sections and determine primary and secondary questions.
Here are some sample questions:
– What motivated you to find a solution to this problem?
– Please tell us about your process of looking for the product to solve the problem.
– What features were important to you when making the decision to purchase the product?
– Have you faced any difficulties when using the product?
Useful tip: When conducting the interview, ask follow-up questions to get to their motives and the context of the environment in which the respondent was at the moment of making a decision. This approach is called the 5 whys technique. It has been observed that, on average, a chain of 5 follow-up questions is needed to establish the cause-and-effect relationships that influenced the respondent’s decisions. It’s beneficial to consider potential follow-up questions for key questions in the questionnaire beforehand.
Here are the examples of 5 whys technique questions:
– In what situations do you use the chatbot?
– Please give a recent example of when you felt the need to use a chatbot.
– What task did you set for the chatbot?
– How did you feel after receiving the chatbot’s solution?
– Did the chatbot solve the task successfully?
4. Interviewing Respondents
When applying the JTBD approach, only live interviews are conducted. Specialized online platforms for respondent search and interviewing are very convenient for this. For instance, in our company, we mostly use the UserTesting platform.
When carrying out an interview, you can follow this advice:
– Start by establishing a connection with the interviewee (tell them a bit about yourself and the plan for the interview, make small talk, etc.).
– Avoid using yes-no questions.
– Be flexible during the conversation and, when necessary and appropriate, change the planned order of the sections.
Useful tip: How many interviews are enough to get to the necessary conclusions? At one point you’ll notice that with each interview, the amount of new information you receive is dramatically decreasing. This will signal that a main portion of the answers to the questionnaire has been collected. In my experience with our projects, this point is usually reached after 30–50 interviews. But it is up to you to decide at what point you have enough data and the research process can be stopped. It’s worth mentioning that accumulating repeated information from different respondents also lets you evaluate the importance of this or that aspect for your target audience.
5. Collecting Artifacts
Artifacts in the JBTD method are the information you extract from respondents’ responses and can be used to enhance your product.
Artifact examples:
– I change my alarm sound when I get so used to it that I don’t hear it in the morning.
– It’s great when people notice that I have an unusual ringtone!
– This app has so many features that it’s hard to figure out how to use it!
Artifacts can be collected either during the interview or afterward.
In the first case, it’s convenient to conduct an interview with a partner, where one person is carrying the conversation while the other is taking artifact notes. Another benefit of working with a partner is that user responses stimulate different ideas in the two people. This way, the same interview analyzed by two different people may generate more artifacts.
In the second case, when you note artifacts after the interview, it’s convenient to have the transcribed conversation (the audio or video recording transformed into text) close at hand. As an example, the online platform UserTesting I’ve mentioned before has such a function. In the generated text, you can highlight specific words and phrases of the users as artifacts. These may help you in promoting your product — they can be used in ad creatives, slogans, app content, etc.
6. Creating Canvases
Canvases are templates for artifact analysis and systematization. A separate canvas is created for each job and includes the responses from different respondents related to this particular job.
Numerous canvas templates are available on the web. Below is a canvas we develop for our projects with an explanation of each block. You can download it here and use it for your projects. It’s also convenient to work with canvases using the Miro service, where you can place canvas template on a virtual board and use stickers to place artifacts on it.
When distributing artifacts across the semantic blocks of canvases, the researcher reevaluates the importance of each artifact for the project and already can formulate hypotheses based on them.
7. Creating a Value Matrix
The next stage of working with the gathered artifacts is creating a value matrix. This matrix will form the foundation for an upcoming brainstorming session, during which the project team will generate ideas for product improvement.
The value matrix is a grid table. The header row lists the driver artifacts collected from the canvases, while the header column lists the barrier artifacts. The order in which these artifacts are listed is random.
During the brainstorming session, solution ideas will be generated at the intersection of each driver and each barrier. These ideas should help leverage the driver to overcome the barrier in order for the product to perform the necessary user’s job. Therefore, artifacts with similar meanings should be combined into a single cell, as solutions to similar artifacts will be common.
Useful tip: If artifacts were noted as user quotes while interviewing and adding them to the canvases, we continue to use this format in the value matrix. This approach allows for a better understanding of the respondent who mentioned the artifact.
For convenience during the brainstorming session, you can give a short general title next to each cell with artifacts in the “Barriers” column or the “Drivers” row reflecting the essence of this artifact or group of artifacts.
8. Brainstorming Using a Value Matrix
For a brainstorming session, we invite those ready to engage creatively and propose innovative ideas. It will be more effective if these people are part of the project team, already familiar with the product we’re researching. From our experience, a group of 5–8 people is optimal.
During the session, everyone should feel free to suggest any ideas. We don’t criticize ideas during this time.
Sometimes, the team may not have a solution to every intersection of a driver with a barrier, or the previous ideas may solve tasks for new intersections. This is a normal part of the process. Simply proceed to the next cell.
Useful tip: It’s helpful if the person who did the interviews shares the main points from them with the project team. Their presence throughout the brainstorming session is also beneficial. If necessary, they can give more information about the context of the interviews from which the artifacts for the value matrix were collected.
9. Working with Ideas
Further work with the ideas from the Value Matrix is done by the project team. The team evaluates and prioritizes ideas to decide whether they deserve to be implemented (this topic is extensive and worthy of a separate article). JTBD is qualitative, not quantitative, research. That’s why before the final decision on the implementation, ideas should be tested if necessary.
JTBD also allows you to review the team’s current list of tasks and adjust their priority if needed.
To sum it up
The JTBD approach enables you to better understand users, test existing hypotheses, and explore new ideas for product development in various areas: the content, the product, the marketing activities, etc.
In-depth interviews allow you to hear live feedback from potential users. And the methodology helps you effectively translate the information you receive into ideas.
I hope this article has helped you appreciate the usefulness of the JTBD approach. Now you know how to use it in practice, and if you have any questions, feel free to ask them in the comments. I wish you success with your projects!
More recommendations from AIBY experts in the next articles. Follow us for more!
