How we do JTBD research at GitLab (A Playbook)

GitLab follows a series of steps and exercises to discover and develop job canvases from basic assumptions all the way to validated and ranked Outcomes and opportunities.

How we conduct JTBD research at GitLab follows a process closely mirroring the JTBD Playbook by Jim Kalbach. The playbook takes teams from basic assumptions all the way to a list of ranked Outcomes that your customers care about the most. Going through this process can be hugely beneficial for a team. These handbook pages and the resources they link to provide you with a solid foundation of JTBD knowledge and a self-serve way to conduct JTBD research collaboratively with your counterparts. However, if you have access to a UX Researcher or someone knowledgeable in JTBD practice, it’s a good idea to have that person as the DRI to help drive this effort for your team.

Use this playbook to complete your JTBD canvases. For more information on each part of the canvas, refer to anatomy of a JTBD canvas.

This page covers the how of Jobs to be Done. For more on what Jobs to be Done is, head over to the main page.

A Note on Time, Confidence, and Validation

The time it takes to run through the entire playbook process can vary substantially, depending on factors such as how many jobs you’re examining, the complexity and variance within those jobs, and your team’s familiarity with the JTBD process. If you’re planning to go through the whole playbook, we recommend budgeting at least one full quarter to get through it.

There are some variables at play here. For instance, in cases where you simply want to identify Main Jobs and sketch out a complete assumptive JTBD canvas(JTBD Canvas Template), no validation is required, and thus the time needed is much shorter. Creating an assumptive canvas can be greatly beneficial as a way to visualize what you know (or think you know) about a particular job, and to get alignment from your team on all the aspects of the job. This can be especially useful where, due to previous research, you have a high level of knowledge about your domain area and how that is reflected in your JTBD. Having said that, high knowledge doesn’t exempt the validation requirements for the latter stages of the JTBD playbook – you will need to complete interviews with your Job Performers to ensure your canvas is accurate, and to understand a Job Performer’s motivations enough to write accurate Outcome statements. When rating your confidence level of your JTBD Canvas, consider the following guidance:

  • High Confidence/Verified: JTBD Canvas has been verified through Job Performer JTBD interviews.
  • Med Confidence/Somewhat Assumptive: Assumptive JTBD Canvas, created based on high level of knowledge of your domain area using existing user research.
  • Low Confidence/Assumptive: Assumptive JTBD Canvas, created based on knowledge of domain area without user research.

With that in mind, the process as defined here will begin with a partial assumptive canvas the way that Jim Kalbach works when leading workhops on the JTBD process.

Open a JTBD issue

If your team wants to conduct in-depth Jobs to be Done research (going through the whole playbook), create an issue in the UX Research Project and use the discover-jtbd issue template. Fill out the issue as best as you can, then tag the UX Researcher for your stage (or whoever will be driving the research). If you need a DRI to drive JTBD research for your team, head over to the #ux_research Slack channel to see if anyone is available to help. Once you have an issue and a DRI, you’re ready to roll!

Workshop 1: Scope Definition

Step 1: Start an assumptive JTBD Canvas to define your Domain, Job Performer, and Main Job

Gather your counterparts together in a sync workshop that should take about an hour of sync time to complete. At a minimum, this team should be comprised of the Product Design DRI and Product Manager. Engineering counterparts are a great addition to round out the perspectives and points of view, and your Section’s UX Researcher would also be a great addition. You’ll gather together on a Zoom call, focusing on the FigJam JTBD Template that was created earlier.

Exercise 1: Brainstorm potential domains [set timer for 10 minutes]

The first step in the JTBD playbook is to select the Domain area in which you will be working. Sections, Stage Groups, or Categories may be a good starting point for determining which area you want to define a JTBD canvas for. In an open area of the FigJam begin by creating new sticky notes for each of the areas within your Group that you might want to develop a JTBD Canvas in. Consider: where do you want to innovate?

Select a domain [set timer for 5 minutes]

Looking at the sticky notes, briefly discuss each, and select one. Because you are very familiar with your work area, you should be able to hone in on a Domain rather quickly. Copy the Domain into the text area provided at the top of the JTBD Canvas. You can keep the unselected Domain areas in this open area in case you want to reference them later to explore additional JTBD Canvases for different Domain areas.

Exercise 2: Brainstorm Job Performers [set timer for 10 minutes]

Now that you know what Domain area you want to work in, it’s time to zoom in on the Job Performer. Below the area where you were exploring Domains, consider; Who are you innovating for? Who are the actors or stakeholders in our Domain/field? Get as many relevant Job Performers written on individual sticky notes as possible.

Select a Job Performer via Note & Vote [set timer for 5 minutes]

To help you determine which Job Performer you want to focus on for this Job Canvas, consider the following questions: Where is there a need in our market? Who are we passionate about serving? What’s relevant or core to our business strategy?

  • Provide each workshopper with some “Target” stickers from FigJam’s sticker set.
    • If < 3 participants: Each participant gets 2 stickers to vote with. You can vote on the same item.
    • If > 3 participants: Each participant gets 1 sticker to vote with.
  • Move the selected Job Performer to the Job Canvas’s Job Performer section.
  • You can keep the unselected Job Performers in this open area if you want to reference them later to explore additional JTBD Canvases for different Job Performers.

Exercise 3: Brainstorm Main Jobs [set timer for 15 minutes]

Next, select the Main Job you want to focus on for this JTBD Canvas. The Job Performer and Main Job are closely related and will ultimately be the focus of the rest of this canvas.

Again, in an open area of the FigJam, consider the following: What are all of the Jobs our selected Job Performer is trying to get done in our Domain/Field? List them all on individual sticky notes. Consider what the primary task or objective is that the Job Performer is trying to accomplish (get done) or an activity they’re trying to complete.

Select a Main Job via Note & Vote [set timer for 5 minutes]

To help you determine which Main Job you want to focus on for this Job Performer’s Job Canvas, ask: What’s worth going after? What are we passionate about? Which can we best support? What’s relevant or core to our business/strategy? What is the primary task or objective that the Job Performer is trying to accomplish (get done)?

  • Provide each workshopper with some “Target” stickers from FigJam’s sticker set.
    • If < 3 participants: Each participant gets 2 stickers to vote with. You can vote on the same item.
    • If > 3 participants: Each participant gets 1 sticker to vote with.
  • Move the selected Main Job to the JTBD Canvas’s Main Job section
  • Move the unselected Main Jobs to the Related Jobs section as these can be investigated in separate JTBD Canvases/research later.

Exercise 4: Brainstorm Aspirations [set timer for 15 minutes]

Finally, this will be the last exercise for this workshop to consider what does our Job Performer aspire to become after completing this Main Job.

One last time, in an open area of the FigJam, consider: What are all of the Aspirations our selected Job Performer is wanting to become after achieving our Main Job? What are they dreaming about becoming when they achieve or complete the Main Job? List the all of the Aspirations you can think of on individual sticky notes. Continue to ask Why to ensure you’re at the highest, aspirational level.

Select an Aspiration via Note & Vote [set timer for 5 minutes]

To help you determine which Aspirations you want to focus on for this JP’s Job Canvas, ask: What’s worth going after? What are we passionate about? Which can we best support? What’s relevant or core to our business/strategy?

  • Provide each workshopper with some “Target” stickers from FigJam’s sticker set.
    • If < 3 participants: Each participant gets 6 stickers to vote with. You can vote on the same item.
    • If > 3 participants: Each participant gets 3 sticker to vote with.
  • Select the top 3 voted Aspirations and move them to the JTBD Canvas’s Aspirations ssection leaving the unselected Aspirations behind for consideraton later, if necessary.

Pre-Workshop 2: Investigation Interviews & Interview Synthesis

The goal of this phase is to understand the Main Job by conducting qualitative interviews with representative Job Performers. This will help us understand and fill in the bottom half of the JTBD Canvas: Job Map, the Emotional and Social aspects, the Outcomes, and the Job Differentiators.

Step 1: Conduct Interviews

Recruit 5-10 external participants representing the selected Job Performer of the Main Job you’re investigating. If you see a lot of variance in jobs, talk to more participants until patterns start to become evident.

We recommend booking 60 minute interviews and having both a moderator and a notetaker. The notetaker is responsible for filling out the Job Steps on the JTBD Interview Participant Canvases during the interview. The canvas will be shown to your participant toward the end of the interview to ensure that we’ve interpreted what we heard accurately.

The goal of the interview is to answer the following questions:

Helpful prompts can be found in the Figjam template interview guide.

Step 2: Synthesize and Complete the JTBD Canvas

After interviews are complete, sythesize all the information together to complete your JTBD Canvas. Include elements you hear most often and discard outliers, tangents, or uncommen circumstances.

When possible, collaborate with your Section’s UX Researcher as experience in qualitative data synthesis is helpful.

Exercise 1. Complete the Job Map

If possible, it is recommended to have both a moderator and a notetaker during each interview. This will allow you to create a Job Map for each session and get feedback from the participant.

  1. Start with a blank slate in Figjam and capture every step the participant mentions using a sticky. Try to maintain chronological order and don’t worry about other aspects of your JTBD, such as job stages or small jobs.
    • Note: If you are unable to include a notetaker during your interview sessions, perform this step afterwards by reviewing your notes and videos.
  2. Once all Job Steps have been discussed, walk through the Job Map with the interviewee for feedback and verification. Make adjustments as needed.
    • Note: If you are not creating the Job Map during the interview with the help of a notetaker, skip this step.
  3. After the interview, rewrite each Job Step to ensure it is functional and follows guidance for how to write a Job Step.

After all interviews:

  1. Look for commonalities among all Job Maps
  2. Create a unified Job Map. Utilize the category labels Beginning, Middle, and End.
  3. Move sticky notes into the category that aligns most closely with the timing the performer would complete each step. Remove duplicates along the way.
  4. Organize sticky notes from each category into vertical stacks. These are related steps the Job Performer would get done, moving vertically down the stack (top to bottom), before moving on to the next stacked step.
  5. Above each vertical stack, add a sticky with stage names or modes of operation. These represent high-level, single-word verbs that describe the collection of steps below it.

Exercise 2: Synthesize Outcomes

Continuing with your interview syntheses, move on to gathering Outcome Statements.

  1. Parse your notes looking for references to how the Job Performer would measure the success of completing the Main Job. Add a sticky for every potential Outcome. Remove duplicates along the way.
  2. Rewrite each Outcome Statement to ensure it follows guidance for properly formatted Outcome statements.

Note: Your Main Job can have between 50 - 100 Outcome Statements which will be validated through an Outcome Opportunity Score survey. This survey can be done in parallel with the rest of your JTBD Canvas work.

Exercies 3: Synthesize Emotional and Social Aspects

Emotional Aspects

Now you will gather together all of the Emotional Aspect Statements from your interviews and then rewrite them into properly formatted Emotional Aspect Statements. Parse your notes looking for references to emotion-like statements that are referencing how the Job Performer feels while doing the Main Job. Add a sticky for every potential emotion you find. Organize them to ensure there aren’t any repeated ideas, merging duplicates as you go.

Finally, rewrite each potential Emotion Aspect Statement converting them into properly formatted Emotion Statements. Check the handbook section for more on how to construct an Emotional Aspect Statement.

Social Aspects

Next you will gather together all of the Social Aspect Statements from your interviews and then rewrite them into properly formatted Social Aspect Statements. Parse your notes looking for references to Social-like statements that are referencing how the Job Performer wants to be perceived while doing the Main Job. Add a sticky for every potential Social/perception you find. Organize them to ensure there aren’t any repeated ideas, merging duplicates as you go.

Finally, rewrite each potential Social statement converting them into properly formatted Social Statements. Check the handbook section for more on how to construct an Emotional Aspect Statement.

Exercise 4: Synthesize Job Differentiators

Now for the final step before your second and last workshop. Gather together all of the Job Differentiators from your interviews and then rewrite them into properly formatted Job Differentiators. Parse you notes looking for references to differentiator-like statements that are referencing any key factors that make a difference on how the Job Performer will get the Main Job done. Look for specific factors that make a difference. Add a sticky for every potential Job Differentiator you find. Organize them to ensure there aren’t any repeated ideas, merging duplicates as you go.

Finally, rewrite each potential Job Differentiators converting them into properly formatted Job Differentiators. Check the handbook section for more on how to construct an Emotional Aspect Statement.

Workshop 2: Map and Prioritize

Step 1. Select Important Job Step(s), Emotional & Social Aspects, and Job Differentiators

Once again you will need to gather your counterparts together in a sync workshop that should take about an hour of sync time to complete. Again, at a minimum, this team should be comprised of the Product Design DRI and Product Manager. Engineering counterparts are a great addition to round out the perspectives and points of view and your Section’s UX Researcher would be a great addition as well. You’ll gather together on a Zoom call, focusing on the FigJam JTBD Canvas you have been working on.

Exercise 1: Review Job Map Select a Job Step(s) [set timer for 15 minutes]

Share the Job Map with your team, read through it to ensure everyone understands the steps necessary to get the Main Job done and to confirm that it’s been accurately arranged and phrased. Make any adjustments as necessary, but be sure you’re referencing the interview data and not going off of gut instinct. Don’t forget, the interviews were performed for a reason.

Select a Job Step(s) via Note & Vote [set timer for 10 minutes]

To help you determine which areas of the Job Map are the most important areas to focus on to affect change for this JobPerformer’s JTBD Canvas, ask: What’s the most important step that’s the hardest to get done? What’s the most critical step?

  • Provide each workshopper with some “Target” stickers from FigJam’s sticker set.
    • If < 3 participants: Each participant gets 2 stickers to vote with. You can vote on the same item as many times as you want.
    • If > 3 participants: Each participant gets 1 stickers to vote with.
  • Discuss the voted upon steps, discussing why each person voted on their step. The goal is to confirm that the team agrees/aligns on a particular step to work on.
    • It is likely that you’ll have a few different steps voted on. Save these for future construction of this JTBD Canvas’s Job Stories.
  • Select and highlight, circle, or outline the top voted Job Step that you are going to innovate on.
  • Copy your Job Map over to your canvas on the right, including the voted on and circled top voted items.

Exercise 2: Select Emotional Aspect Statements [set timer for 5 minutes]

Share the Emotional Aspect stickies with your team, have everyone read through and understand them.

Select Emotional Aspect Statements via Note & Vote [set timer for 5 minutes]

To help you determine which Emotional Aspect statements are the most important areas to focus on to affect change for this JobPerformer’s JTBD Canvas, ask: What’s worth going after? What are we passionate about? Which can we best support? What’s relevant or core to our business/strategy?

  • Provide each workshopper with some “Target” stickers from FigJam’s sticker set.
    • If < 3 participants: Each participant gets 6 stickers to vote with. You can vote on the same item as many times as you want.
    • If > 3 participants: Each participant gets 3 stickers to vote with.
  • Arrange the voted on stickies in a pyramid with the top voted stickies at the top and lowest at the bottom.
  • Discuss the top voted upon items and confirm that the team agrees/aligns on the top voted Emotional Aspect Statements and move them to the appropriate space on the JTBD Canvas.

Exercise 3: Select Social Aspect Statements [set timer for 10 minutes]

Repeat the same process you just performed for the Emotional Aspect Statements for the Social Aspect Statements. Move the top voted Social Aspect Statements to the appropriate space on the JTBD Canvas.

Exercise 4: Select Job Differentiator Statements [set timer for 5 minutes]

Repeat the same process you just performed for the Emotional and Social Aspect Statements for the Job Differentiator Statements. Move the top voted Job Differentiator Statements to the appropriate space on the JTBD Canvas.

🎉🥳 Congratultations you have successuflly created a validated JTBD Canvas! 🥳🎉

Outcome Opportunity Scores Survey

Now that you have your JTBD Canvas filled in completely, and you have decided on the most important areas to focus on for each section but the Outcome Statements, you need to run an Outcome Opportunity Scores Survey to determine which of these are the most important to your Job Performer for this Main Job.

To do this you need to construct an Outcome Opportunity Score survey. It is recommended that you work with your Section’s UX Researcher when constructing and delivering this survey.

This survey asks two questions about each of your Outcome Statements:

  1. How important is this (Outcome) to you?
  2. How well is this (Outcome) currently being satisfied?

It’s important that your survey participants are reprensetitive of your Job Perfomer so you’ll want to be careful to screen out any participants who don’t perform the Main Job you’re researching. Try to get close to 50 qualified responses to ensure a reasonable amount of confidence in your scoring.

There is a template in Qualtrics to help you get started. (Under Library->UX Research & Product->Templates->Opportunity Score Template).

Note: Remember to recruit both GitLab and non-GitLab users, and be able to easily segment them for your analysis.

At GitLab, we use a 10 point scale, labeled at each end as ‘very low’ (1) and ‘very high’ (10).

Using the scores for importance and satisfaction, we can calculate an opportunity score:

JTBD Opportunity Score

  1. The importance score minus the satisfaction score is your ‘satisfaction gap’.
  2. Add the satisfaction gap to your original importance score to get the final opportunity score.

This spreadsheet template offers a step by step guide to exporting data from a Qualtrics survey and generating a ranked list of Outcome statements with opportunity scores.

Opportunity scores run from 0-20. Generally, anything above a 10 is worth looking at closely, and anything over 15 is a critical opportunity that needs to be addressed as soon as possible.

If all the scores on your survey are less than 10, that’s great - but remember, any score above a zero indicates some room for improvement, so it’s still wise to look at your top scores. Conversely, if you have dozens of Outcome statements above a 10, prioritize those with the highest scores.

Opportunity Score Severity
16-20 Act immeditately - house is on fire!
10-15 Prioritize as soon as possible
7-9 Keep on eye on
6 or below Probably not a priority right now

Generate Job Stories

Now that you have a completed JTBD Canvas you can use it to create Job Stories. Job Stories help synthesize and summarize all of the great information you just put together on your JTBD Canvas. You might end up creating between 3 - 5 Job Stories to cover the data and insights you gathered and voted on as being the most important aspect of your JTBD Canvas. Bear in mind that good Job Stories describe the pain points that you’re going after and help you empathize with the Job Performer.

To learn more about what Job Stories are, how to write them, and how you can use them to generate design solutions check the handbook section.

The Job Story Fomula

When I ___________ [am at this Job Step] + [under these conditions-Job Differentiators], I want ____________ [this New Ability, customer imperative or demand the JP has on the solution], So I can __________ [reach these Outcomes] + [and have these Emotional/Social Aspects].