JTBD Research How-To's: Avoiding the Traps of Poor Foundations

Discover the four critical elements that can make or break your JTBD research foundation

Mar 12, 2025

Table of Contents

The longer I’ve been in this space, the more I’ve seen really horrible foundations for JTBD research. I’m a big proponent of having a comprehensive research plan; especially when you get one shot at a survey. I have a mini-course on this where you can learn more. It’s located in my community. You can join my FREE community and earn free access to everything by accumulating a few engagement points.

Elements of a poor foundation

A JTBD value model that is not collectively focused on a singular objective
A JTBD value model that is not framed at an actionable level of abstraction on whole
Success metrics that do not relate to the functional objective (i.e. consumption metrics, financial metrics, be goals, emotions, etc.)
Success metrics that are not pre-determined to be actionable should they be determined as an opportunity, i.e., they are underserved, or overserved

Singular Objective

The prompts I crafted for developing JTBD value models are designed to ensure that Jobs, Steps, and Metrics all fit together like a glove. They’re run through a test-fit formula. With a verbose output (reasoning) you’ll see a final check. Here’s an example of that for the following inputs:

Job: restoring blood flow

Context: surgically

End User: heart surgeon

Fidelity: high

Start Point: perform surgical procedure

End Point: complete the surgical procedure

"As a heart surgeon who is restoring blood flow surgically, you need to conduct pre-operative time-out, confirm availability of surgical tools and materials, position the patient, prepare the surgical field, verify patient’s physiological readiness, make the initial incision, expose the target area, isolate the affected vessel, perform the intervention, verify restoration of blood flow, monitor vital signs, monitor for complications, manage intraoperative complications, adapt surgical strategy, close the surgical site, and ensure post-operative stability."

As additional components and levels of the model are constructed (approaches, success metrics, potential root causes), deeper test-fits are checked as well. The entire model is cohesive based on the scoping inputs you provide. While you can manually fiddle after-the-fact, I’ve seen far too much practitioner bias, or stakeholder influence (that amounts to fiddling) on the front-end for my taste. This results in an incoherent, incongruent model.

Let’s be honest, both consultants and client stakeholders can have a detrimental impact on model development. It’s not a lack of experience, it’s a lack of commitment in many cases. Or, it’s a financial decision, i.e., need cash flow now so ‘let customer break the methodology’, etc. This seems harsh, but if a client is paying you 6-figures to help them make 7+-figure strategic product or business model decisions, you need to provide insights that have the highest probability of being both accurate and actionable. If not …

…your client does not know what to do with the results and your PowerPoint deck gets lost in a sub-folder somewhere.

Actionable Level of Abstraction

Another problem I’ve seen is research models that are not in the abstraction sweet spot. A simple way to think of this is using a pyramid. At the top we have the JTBD “Be Goal” crowd. They want you to believe that you can innovate by understanding the feelings or even the motivations of people. What do they want to become?

Going a level down gets us closer to innovation reality. Wishful Functional is a space where brands will focus on insights in areas that they are not capable of fulfilling given their current enterprise capabilities in the foreseeable future. How much research investment do you want to make in areas where you have high uncertainty about your ability to execute?

The correct place to focus your attention is closer to the bottom of the pyramid.

There we find functional models that can be studied where you have confidence that you can take action if you find value gaps in the market. Those gaps align well with your ability to execute / take action in the foreseeable future with your current capabilities, easily developed capabilities, or easily acquired capabilities. This is where you play if you want something that is a breakthrough or even disruptive.

At the bottom we have consumption jobs. These are the 17 customer journeys (and related service pathways) where you can either improve the experience or improve the service. These are generally going to be incremental improvements to an existing business model and offering.

You may also find opportunities to develop new horizontal business models that cater to customer groups beyond your current market by providing a service common the them all but have historically been performed within the category, or in-house.

What I see in every faction of JTBD is either wishful functional modeling, incoherent modeling, or aspirational modeling. None of these are useful for anything beyond conversation. So I would not invest in them.

If you need a second opinion, you can always ping me.

Unrelated Success Metrics

I’ve seen this many times, where a stakeholder pushes to have you incorporate metrics into the core model that having nothing to do with the model. Being fair, I’ve seen practitioners do this as well, due to inexperience I guess. It’s to hard to know the reason since we enter this domain having a very clear set instructions with regard to the methodology.

For example, emotional and social statements should not be a customer success metric. Financial metrics — which tend to relate to total cost of ownership — should not be included in the model because many end users are the wrong audience to answer them … and they have little do with getting the functional job done. Consumption metrics also creep into the job map which is completely inappropriate and irrelevant. We have a place to address them separately if you have questions that need to be answered.

Here’s how we should deal with these things:

Emotional and Social jobs statements have their own section of the survey which can then be correlated to insights in the job map.
Financial Metrics have their own section for consumer surveys, but require a separate survey in most B2B studies.
Consumption Metrics also have their own section of the survey. This can be useful information to capture … just not inside a function core job map that is not about a customer journey.
Experiments are something I like to do. Also, stakeholders will have suggestions. I hide these inside the Related Jobs section. As long as the audience you are surveying is the right person to answer the question — and you have room in the survey — go ahead and ask those questions

Unconfirmed Success Metrics

One of the most obvious mistakes that is made is handing a final model to your client and asking them if it looks good. A casual acceptance is not good enough. Here’s what you need to do in order to be successful.

Go through every single success metric with your stakeholders (sponsor, design, product) and make sure that they all agree that they can at least imagine a solution for the metric should it be underserved. If they can’t agree on this, then either your model is in the Wishful Functional or Aspiration abstraction, or the success metric is bogus and should be removed.
Even after you gain alignment on this, you need to make sure an ideal respondent can pass a cognitive review of the survey. Are they able to understand the question? This is a process you can look up called cognitive interviews. Just Grok it!

These two steps ensure that your client knows they can take action on anything before you run a survey and that the people you want to take the survey can interpret your questions the way you intend. Anything short of this is a wasted investment and could result in your client having invested 6 figures for a useless PowerPoint, and maybe a spreadsheet that they don’t know how to use.

Mike Boysen - www.pjtbd.com
Why fail fast when you can succeed the first time?

🚨 Join my FREE community for all of my AI Prompts, JTBD Courses and weekly AMA office hours: https://pjtbd.com/join

📆 Book an appointment: https://pjtbd.com/book-mike

🆓 Free stuff in my Notion database:

https://pjtbd.notion.site