Published April 12, 2024

How JTBD Research with AI Revolutionizes Product Development

12 min read

In product development and market innovation, understanding the ‘why’ behind consumer choices is as critical as the ‘what’. This is where Jobs-to-be-Done (JTBD) research steps into the spotlight. It’s a methodology heralded for its ability to get to the heart of the customer’s needs – identifying why people ‘hire’ products or services to accomplish their goals. 

We recently came across a way to speed up the process of doing JTBD research using Jim Kalbach’s experimentation with ChatGPT to fill out the JTBD canvas. He emphasizes that these models are experimental, and AI should complement field research, not a replacement—it’s a tool to accelerate and deepen your research while maintaining essential customer connections. Check out his guide on how to use AI.

For me, using AI is a way to accelerate research so you can do more it and dig deeper where needed. And as always, I’m not suggesting AI should completely replace field research. It would be hard for me to imagine becoming more customer-centric across a team or org without having direct contact with the people you serve. Customer interviews aren’t just about data collection: they build team connections and foster empathy for customers, among other things.

Avatar of the person that wrote the post

Jim Kalbach

Chief Evangelist at Mural

We shouldn’t think of AI as a buzzword; it’s a potent tool changing how we conduct JTBD research. It’s accelerating the process, allowing us to dive deeper and scale new heights in understanding customer behavior. But how can you effectively integrate AI into your JTBD research? This post isn’t just about the ‘how’ – it’s about transforming the ‘how’ into a resounding success.

The JTBD Framework: A Primer

Before we delve into the mechanics of AI in JTBD research, let’s revisit the fundamentals. The JTBD framework is a lens through which we view our customers not just as consumers but as individuals trying to get jobs done in their lives. These jobs could be functional, like getting from point A to B, or emotional, like feeling secure in an investment.

This is where Jobs-to-be-Done (JTBD) research steps into the spotlight.

JTBD framework breaks down this concept into several key components:

  • Job Performer: Who’s getting the job done? This could be a specific demographic or an individual in a particular role.
  • Target Job: What’s the primary job the product or service is being ‘hired’ for? This is the crux of the matter – the problem that needs solving.
  • Related Jobs: These additional jobs orbit around the target job. Think of them as the supporting cast that amplifies the value of solving the primary job.
  • Aspirations: This is about the dreams and goals of the job performer. What do they hope to achieve by completing the target job?
  • Job Steps: The stages involved in completing the job, which can typically be broken down into the beginning, middle, and end.
  • Emotional & Social Aspects: These elements capture the feelings and societal implications tied to the job.
  • Outcomes: What does success look like from the perspective of the job performer?
  • Job Differentiators: These are the unique selling points that make one solution stand out from another in the eyes of the job performer.

Understanding these components is the first step in JTBD research, and it’s where AI begins to show its value.

The AI Advantage in JTBD Research

Now, let’s talk about supercharging your JTBD research with AI. Jim Kalbach’s experimentation with AI tools like ChatGPT to fill out the JTBD canvas is a testament to the potential of AI in this field. He illustrates that AI can complement field research, not a replacement. It’s about using AI to speed up the process and enrich your findings, all while keeping that crucial customer connection intact.

AI can interact with the various elements of the JTBD framework, refining outcomes through an iterative process. For instance, defining job performers can be as simple as asking an AI to list relevant actors in a field and provide descriptions of their roles. This can range from home buyers to real estate agents in the property market, each with a unique job to be done.

The beauty of AI in this context is its ability to handle large volumes of data, detect patterns, and draw insights at a scale and speed that’s simply unfeasible for human researchers alone. It’s a force multiplier for your JTBD research, providing a richer, more nuanced understanding of the customer jobs landscape.

In the following sections, we’ll walk through how to deploy AI at each step of the JTBD framework, from identifying job performers to distilling job differentiators. We’ll showcase tools and techniques, share best practices, and look at real-world cases where AI-driven JTBD research has significantly impacted.

Filling out each section of the JTBD framework

For each of the sections, Jim shows how to fill out the Jobs to Be Done framework using ChatGPT.

The Jobs to Be Done (JTBD) framework is a methodical approach that allows businesses to identify and understand the core tasks their customers need to accomplish. At the outset, the ‘Job Performer’ is pinpointed, defining the individual or role that the product or service is intended for. The ‘Target Job’ section articulates the main objective or problem the product or service is ‘hired’ to solve for the job performer. Additionally, ‘Related Jobs’ encompass the supplementary or subsequent tasks that are part of achieving the main job, while ‘Aspirations’ delve into the higher-level goals or desired states of being that the job performer aims to attain through the job​​.

The framework breaks down the ‘Job Steps’ into beginning, middle, and end to structure the progression of completing the job. It also captures ‘Emotional & Social Aspects’ to reflect the emotional journey and social context of the job performer. ‘Outcomes’ focus on the desired results or success metrics from the perspective of the job performer. Lastly, ‘Job Differentiators’ highlight the unique factors that make one solution stand out over another in the eyes of the job performer, offering a competitive edge in the marketplace​

JTBD Research AI Example with Helio

In the following sections, we’ll explore how to use AI to create your initial research, and Helio to validate your assumptions.

Job Performer: Who’s On Stage in the Jobs Theater?

Imagine a bustling stage – the marketplace. The performers here are as varied as the roles in any grand play, each with a script shaped by their objectives. In the real estate spectacle, our cast includes home buyers, with their dreams of a hearth to call their own, and real estate agents, the directors who guide the narrative towards a happy ending. Yet, the ensemble doesn’t end there; it spans inspectors, who ensure the stage is set safely, to loan officers, the financiers of dreams, each playing a part in the grand production of buying and selling homes.


In software development, the plot is different but equally complex. Project managers orchestrate the production, developers craft the scenes, and quality assurance testers keep the performance flawless. Each performer has a role, a purpose driving their actions on this stage.

Through the lens of AI, we can not only list these performers but dive into their scripts, understanding their lines and cues in the market play. This adds precision to our understanding and ensures our product is the perfect stage for them to perform their jobs.

Here’s an example prompt:

PROMPT 1: Define the job performers
Who are job performers in my field?

I’m working with the jobs to be done (JTBD) framework and would like to find job performers in the field of software development

The job performers are the key actors or roles that have objectives to accomplish within that field.

For instance, in the field of buying and selling homes, job performers would include, but are not limited to: home buyers, home sellers, real estate agents, lawyers, inspectors, loan officers, mortgage lenders, titling agents, and neighbors.

Please list [[n]] job performers in bold followed by a one sentence description of their role in that field. Each job performer should be singular in nature, with no “ANDs.”

We put these job performer prompts to the test by entering them into ChatGPT ourselves. We focused on a scenario that we could put to the test with an audience of our own, so we asked the AI software to give us the top 5 job performers in the field of software development:

give us the top 5 job performers in the field of software development.

Now with the initial ideas provided by our AI friend, we turned to our audience of Software Developers and IT Professionals in the US to see if their answers revealed similar insights.

We started by reaching a larger pool of participants that work as tech professionals, and then filtering the data down to those developers and tech engineers that we want to hear from:

Filtering the data down to those developers and tech engineers.

As we went through the responses provided by our audience, we were able to compare and contrast with the answers produced by ChatGPT.

View the Helio Test Report

Target Job: The Star of the Show

Every performance has its lead, and in the world of Jobs-to-be-Done, it’s the ‘Target Job’. It’s the main act, the central problem our product is hired to solve. Whether it’s to provide a home that’s a sanctuary for a family or to streamline a project to its successful delivery, identifying this job is like defining the theme of our play.

AI doesn’t just spotlight the lead; it helps articulate the role in high definition. It ensures that every line spoken resonates with clarity and focus, so the job performer’s needs are met with a standing ovation.

In our scenario of finding the jobs-to-be-done in software development, ChatGPT said that designing software architecture, debugging software issues, and collaborating with team members are top of the list. 

While these jobs are definitely relevant, we found more through our audience of tech professionals:

“The “job” in this context is releasing a new business feature on our website. Releasing the feature means making it available for end customers to see and use it. Launching the same means making it active in a real end user environment.”
Helio Participant, Software Developer (US) 

Developers are often involved in almost the entire life cycle of a feature or product, from concept evaluation to code optimization. Many of our participants spoke to the idea that they are working on a new feature or concept for their company or client, so we focused on the scenario of ‘launching a new feature’ for our follow-up AI prompts.

Related Jobs: The Supporting Cast

No star shines without a supporting cast. These are the ancillary jobs, the side tasks that, while not the main act, are essential to the plot’s progression. 

In home buying, it’s the background checks, the financing, and the negotiations. In software development, it’s the code reviews, the sprint plannings, and the stakeholder updates, as we found from testing our audience of developers:

“Developed, QA, and launch a feature within a certain timeline.”
– Helio Participant, Software Developer (US)

QA was an element mentioned by many of the software developers in our audience, similar to the ‘debug software issues’ job that ChatGPT suggested based on the prompt. However, where ChatGPT listed this as one of the primary jobs for our target audience, most of our developers listed it as a step in the road to the final goal of launching their new feature.

Aspirations: The Dream Behind the Role

Beyond the script lies the dream. The standing ovation, the curtain call, the higher-level goals the performers aspire to. It’s not just about buying a house; it’s about creating a home. It’s not just about delivering software; it’s about innovating solutions.

AI helps us uncover these aspirations, adding a layer of depth to our understanding, ensuring our product doesn’t just meet the basic job requirements and fuels the dreams that drive our performers.

Technical expert as a JTBD.

We were able to confirm some of these aspirations, like the contributions to team success, with our participants in Helio:

“be proud of the work that I helped contribute to the collective efforts of building out this new feature”
– Helio Participants, Software Developer (US)

While ChatGPT was obviously quick to deliver these ‘be’ goals for software engineers, our audience was less interested about their personal aspirations and more keen on delivering on wholesome aspirations for the product:

“Smooth deployment, no bugs or negative impacts on the users and business, improving the system”
– Helio Participant, Software Developer (US)

Terms like ‘smooth’ and ‘accessible’ were common aspirations that developers had for their end product.

Job Steps: The Script in Detail

Every job has its storyline, divided into the beginning, middle, and end. The beginning is the overture, where intentions are set and the journey begins. The middle is the development, where the plot thickens and actions take shape. The end is the resolution, where outcomes are achieved, and the audience is left satisfied.

AI can be our scriptwriter, helping to draft these steps precisely, ensuring that the journey from curtain rise to fall is seamless and impactful.

We want every audience left satisfied.

ChatGPT’s layout of 12 steps in the process are holistic, and certainly match with what our audience of software developers explained.

“Have UI designs ready for me to execute on, add any necessary data to database that is needed for the new features, create APIs to get the data I need, complete coding, have work QA’d, address and fix any bugs or minor changes found during QA/review.”
– Helio Participant, Software Developer (US)

Most developers’ list of job steps topped out around 6-8 items. Those who produced longer lists were more clearly involved in the beginning to end conceptual development of the feature, all the way through its launch and live evaluation, while others simply worked up until a code-ready position.

Emotional & Social Aspects: The Heart and Soul of the Performance

Every performance has an emotional undercurrent, and every market has its social context. It’s the thrill of finding the perfect home, the stress of a project’s deadline, the camaraderie of a team overcoming challenges, or the frustration of unexpected setbacks.

AI helps us tune into these frequencies, identifying the emotional and social melodies that play a crucial role in the overall experience of the job performer.

Identifying the emotional and social melodies that play a crucial role in the overall experience.

While ChatGPT surely produced general aspirations that were shared by our audience of software developers, the missed out on a key emotion that most of our participants reflected:

“I feel confident seeing that what I and my team achieved at the end of the day is actually to the standard of the company and good enough to the users.”
– Helio Participant, Software Developer (US)

‘Confidence’ was the most common term used by developers to describe how they want to feel at the end of a project. This does involve how teamwork makes them feel, as mentioned by ChatGPT, though it’s much more centered on a future prediction of success.

Outcomes: The Encore They Seek

What’s a performance without applause? The outcomes are the applause, the standing ovation job performers seek. It’s the satisfaction of a home well bought, or the pride of a project well delivered.

AI aids us in defining these outcomes, providing the metrics by which we can measure the success of our product, ensuring our performers leave the stage to the sound of cheers and not silence.

The outcomes are the applause, the standing ovation job performers seek.

ChatGPT seems to sing a repetitive tune here, mostly focusing reducing negative results. The terminology of the AI prompt may be a cause here, as the examples provided in the prompt only focus on minimization.

However, our audience of software developers in the US also expressed a desire for maximizing:

“Increase the amount of traffic to the website.”
– Helio Participant, Software Developer (US)

While some developers were focused on minimizing errors, those with a more holistic approach to launching a feature are also focused on how the new feature improves the engagement and traffic on a platform.

Job Differentiators: The Unique Selling Point

What makes our performance worth the ticket in a world of many stages and countless plays? It’s the job differentiators, the unique props, the special effects, the exclusive scenes that our product offers.

With AI, we can spotlight these differentiators, ensuring that when the curtains close, it’s our show that the job performers remember, recommend, and return to, time and again.

In weaving the Jobs-to-be-Done framework into our market strategy, these elements form the narrative that resonates with our audience and elevates our product to be the star of the marketplace stage.

Create narratives that resonate with your audience and elevates your product.

ChatGPT gets very specific here, which is where we see its ultimate value compared to the responses of our participants:

“Clear objectives, effective communication, and collaborative teamwork are the most crucial factors in getting the job done effectively.”
– Helio Participant, Software Developer (US)

Clarity and communication about the goals amongst a team were the most common differentiators that our audience of software developers described. These seem to be the big picture buckets that the specific factors produced by ChatGPT would fall into, such as the in-person or remote nature of the team.

This side-by-side comparison of ChatGPT’s outputs and responses from a live audience show how each system of gathering information can be useful in different scenarios. To optimize the learning process about your audience’s jobs-to-be-done, it would make sense to first see what insight an AI solution can provide into the subject, and then follow up with a Helio survey where you confirm those answers with a relevant user audience.

The Human-AI Partnership in JTBD Research

This is the new frontier of product discovery and UX research – a partnership between human ingenuity and AI’s computational power. The possibilities are vast, and the potential for innovation is immense. AI doesn’t just accelerate JTBD research; it enables a deeper, more nuanced understanding of customer needs and wants.

So, as you embark on your next JTBD research project, consider AI your indispensable wingman. It’s here to help you fly higher, dive deeper, and uncover insights that could be the key to your next breakthrough product or service.

JTBD Research with AI FAQ

What is Jobs-to-be-Done (JTBD) research and why is it important?
Caret signaling that you can click it to open the dropdown

JTBD research is a methodology focused on understanding the underlying reasons and needs behind why consumers hire products or services to accomplish specific goals. It’s crucial because it helps businesses identify what truly motivates their customers, beyond surface-level desires, enabling more targeted and effective product development and innovation.

How can AI accelerate the process of JTBD research?
Caret signaling that you can click it to open the dropdown

AI, particularly tools like ChatGPT, can significantly speed up JTBD research by automating the process of filling out the JTBD framework. This includes identifying job performers, articulating the target job, and understanding related jobs and aspirations. AI complements field research by allowing researchers to conduct more thorough investigations and dig deeper into customer needs, without entirely replacing the invaluable insights gained from direct customer interactions.

What are the key components of the JTBD framework?
Caret signaling that you can click it to open the dropdown

The JTBD framework includes several key components: Job Performer (who is getting the job done), Target Job (the primary job the product or service is hired for), Related Jobs (supporting tasks), Aspirations (the goals the job performer aims to achieve), Job Steps (stages in completing the job), Emotional & Social Aspects (feelings and societal implications), Outcomes (what success looks like), and Job Differentiators (what makes one solution stand out).

How does AI interact with the JTBD framework?
Caret signaling that you can click it to open the dropdown

AI interacts with the JTBD framework by engaging with its various elements, such as defining job performers, refining outcomes through iterative processes, and identifying job differentiators. This interaction enables handling large volumes of data, detecting patterns, and drawing insights at a scale and speed unfeasible for human researchers alone, thereby enriching the understanding of customer behavior and needs.

What are some real-world applications of AI in JTBD research?
Caret signaling that you can click it to open the dropdown

Real-world applications of AI in JTBD research include using tools like ChatGPT to define job performers in specific fields, articulate the target job, and identify related jobs and aspirations. These applications enable businesses to gain a deeper, more nuanced understanding of their customers, streamline the product development process, and ultimately deliver solutions that better meet consumer needs.

Can AI replace direct customer interactions in JTBD research?
Caret signaling that you can click it to open the dropdown

While AI can significantly enhance and accelerate JTBD research, it should not replace direct customer interactions. Customer interviews and field research are essential for building team connections, fostering empathy for customers, and gaining insights that AI alone cannot provide. AI is a tool to complement and extend these interactions, not a substitute.

How can businesses effectively integrate AI into their JTBD research?
Caret signaling that you can click it to open the dropdown

Businesses can effectively integrate AI into their JTBD research by using AI tools to fill out the JTBD framework, starting from identifying job performers to distilling job differentiators. The key is to leverage AI for its ability to process information quickly and identify patterns, while still relying on direct customer feedback and field research to ensure a comprehensive understanding of customer needs and behaviors.

Build something your users truly want