What I learned from doing 100 Jobs-To-Be-Done interviews
A little over two years ago I sat in the 37signals office listening to Jason Fried,Bob Moesta and Chris Spiek talk about Jobs-To-Be-Done (JTBD) as a useful framework for understanding why customers switch from one product to another.
At that point in my career I had commissioned market research in my various marketing roles, but the JTBD approach seemed to be a great way to get at some of the underlying motivations and behaviours that were often hard to pinpoint.
One thing I’ve learned from being a student of anything is the sooner you put it into practice, the more likely you are to retain the material. So, after this first workshop I went home and did 10 interviews on willing friends, just to get my feet wet.
Since then I’ve been able to do another 90 or so interviews with various companies, mostly early and growth stage startups. I’ve done interviews relating to products ranging from enterprise software to consumer apps to coaching. I’ve also spent more time with the ReWired guys and other JTBD practitioners, trying to learn as much as I can.
I thought the 100-interview milestone was a good opportunity to stop and reflect on what I’ve learned so far:
Before you do any JTBD research, start by articulating what you’re hoping to learn
As my colleague Rebecca said so well in her recent Research Design workshop, before you rush off to do some JTBD interviews, you need to ask what questions you are hoping to answer by doing research in the first place. Are you trying to learn why people stop using your product or maybe you want to know what the real competitors are from your customer’s point of view?
Note this does not mean you need to ask the questions directly while doing the interviews. It’s more of a focusing question for yourself and for your organization so you are doing research that provides value and know what areas to probe on more.
There is no shortcut for becoming a good interviewer
The bad news is that most of us are terrible at doing customer interviews, even if we think we are empathetic or have lots of experience talking to customers. The two most common mistakes I’ve witnessed and made myself are: asking leading questions and not being comfortable with silence.
The good news is, you can get better through learning from the experts and lots of practice. There are three different things that helped me get better at doing interviews:
- Sitting in on interviews with people who were better than me. I started by taking some training with Bob and Chris, during which we got to observe them doing interviews. I also sat with other experienced interviewers to observe how they interview. This helped me tremendously.
- Recording my interviews. There are many benefits to recording an interview; one of which is observing your own interview skills. I have often done a “great” interview, only to go back and see where I could have improved. This is normal and part of the process of getting better over time.
- Practice. A lot. Like any other skill, practice makes perfect. There is only so much you can learn passively by observing other people. At some point you have to just put in the time to do the interviews yourself.
There are limitations to what you can get from an interview alone
While most people use interviews to gather JTBD data, the interview format does have its drawbacks. For one, people don’t always recall details that may be important. I also found it challenging to extract details for purchases that were less considered, i.e. a $2 app versus a $10,000 enterprise software purchase. (By the way, the energy around a purchase is not necessarily related to price, it might have more to do with how the person valued that product category. For example, a person might have a high recollection of details around a $40 product related to their health versus a $300 service for their car.)
It’s also more challenging to understand the JTBD for a product that doesn’t exist versus something that people are buying today. JTBD is often talked about in the context of innovation and it’s a useful model for understanding disruption. When you are trying to design products that don’t exist today, doing interviews is less straightforward — you need to find ways to learn about behaviours around products and services that may be today’s substitutes for a future product.
There are other ways to collect data that can be analyzed using a JTBD lens — ethnography, diary studies, generative research, etc…which I haven’t tried myself but have observed others do successfully. I’d love to dig into this more in the future.
Analyzing JTBD data can be as hard as collecting the data
Collecting the JTBD data via interviews or other methods is just the first part of using the framework. But what do you do with all the information? This is where the biggest gap exists in terms of the existing materials on JTBD. Yes there are some models to frame the data, like the Timeline and Four Forces, but there is little out there that shows you how to get from looking at a large amount of data to actually deriving the insights from that data.
I’ve tried both loose methods (using my notes and “smarts” to extract insights) to more formal techniques such as open and closed sorting. There is also software out there that can help you tag and sort transcriptions of interviews.
This is one area where I’d love to see more development, tools and training.
Not all jobs are created equal
People may have multiple “jobs” that they hire a product for, but they are likely not weighted equally. In JTBD terminology this is known as the big hire and little hires. For example, I might hire Uber to make my travel expense reports less of a hassle (big hire) but I might also be hiring it to impress my friends while getting from point A to B (little hire).
The big hire is the one that should get most of your attention but the little hires can help you design interactions, messaging and experiences that can make the difference in your customer choosing to hire your product versus another.
Apply insights, measure, repeat!
This one seems obvious but whenever I do JTBD interviews I make sure that I loop back to the original reason for doing the interviews. It’s not enough to show the insights to a group and have them nod and say “that’s interesting.” JTBD is only useful if it allows you to do something differently than before you had that information.
Whether it’s designing new products, changing product features or refining messaging and marketing, make a list of how you are going to apply what you learned and how you are going to measure the results of these changes. The folks at Intercom have written up a great example of how they did this.
I found through my own experience that sometimes JTBD insights led to changes that had dramatically positive results, but there were also times where the application of our insights didn’t move the needle enough. In the latter case, we had to do further research which we then applied and measured, repeating as needed until we achieved our goals.
JTBD is a useful framework for orienting around behaviours, not features
After 100 JTBD interviews, I would say that it is a valuable framework for looking at markets, products and marketing. Thinking about how and why customers hire products and services has helped me have more of a customer-orientation versus a company-centric orientation. The Four Forces model in particular has been extremely helpful in shedding light on what keeps people from switching to a new product or service despite its bells and whistles or logical advantages. It’s not enough to have a great product. There is so much underlying emotion around purchases, even those that are seemingly mundane, and I have found this to be the one area that most companies overlook.
===
I’d love to hear from others who are using JTBD to see if my experience lines up with yours, or if you have additional comments and insights to share.
