I probably should not admit this – as an innovation professional – but I just finished reading “Contending Against Luck,” Clayton Christensen’s last work on Innovation.
If you aren’t aware, he is the one who came up with the concept of disruptive Innovation over 20 years ago and encapsulated his findings in the seminal book on Innovation, The Innovators Dilemma. For some, this is the bible of Innovation, cited in many PowerPoints, whitepapers, and websites about the right way to innovate.
In that book, he talked about how Innovation grows out of the white spaces, which many companies are blind to as they are focused on what is happening today instead of also hedging their bets on what may happen tomorrow. Since then, he’s mentioned a few times that his original definition of Innovation may have been faulty since so many innovation initiatives fail.
To attempt to remedy this, he developed a new theory of Innovation, which he’s called “Jobs-To-Be-Done.” This is outlined in the book above, and in short, is a way to redefine Innovation along usefulness lines. In the original definition, disruptive innovation may have driven people and companies to try many paths, which may have led nowhere because, in the authors, words – did not map to a “Job to be done.”
In its simplest terms, JTBD theory is that every customer has a “job” that they wish to complete, and they “hire” something to solve that job. Sounds incorrect since “solving” is usually matched with “problem” and not “job,” but in Christensen’s theory, jobs are solved. Let’s talk through the theory: customers have a need (something that they want or need) – which translates to a job (a task which needs to be performed to fulfill that need) -, and if you can do the job, the customer “hires” your product to do the job.
He gives a “milkshake” example – it’s 10 am, and you are hungry as you ate breakfast early. You would think that the need would be hunger – but needs are more complex than that – the need could be hunger, plus a craving for something sweet, plus a sense of losing control as your day may be getting away from you. This need is complex, and the “job description” for this need covers many aspects – personal, professional, emotional, social, etc. This need is not as simple as hunger since many things can slake that hunger.
The customer in the example chooses to have a milkshake, as the milkshake’s “resume” fulfills the “job description” better than a bagel or beef jerky – it deals with the hunger, the sweet tooth, and the loss of control simultaneously. This is why the customer decides to have a milkshake – it “does the job.” In this assumption, the customer job is a very complex set of elements along all of those criteria and more, so many products and services fail – they do not do the job. They may do some of the jobs, or they may do most of the job, but they do not do enough of the job to satisfy the customer. This explains the disconnect between successful and unsuccessful innovations – the successful ones figured out the job to be done and did it, where the others didn’t.
An example of an innovation that solved a job? Popsockets, which were created as headphone holders – became phone holders and stood when given to potential customers. They solved the job of securely holding your phone and as a stand to watch videos on. An innovation that failed: HomeJoy, the Uber for cleaning services, did not solve the job. If you look back on the many failures and successes of innovation initiatives, you can trace them back to the product’s failure or success or service solving the “job.”
As I read this, I realized that this is very similar to the “customer desires” concept that I have been discussing for the last five years now – applied to autonomous processes. In this book, Christensen suggests that to develop successful innovative products and services, you must determine a customer’s “job to be done,” then solve for that job, then provide a product or service to meet that solution. How is this any different from marketing assessments and applying customer feedback? It’s a much deeper dive into the customer’s mind – determining their desire along all of the above parameters.
This is not easy to do – plus at the same time, there are privacy issues with customers revealing the depth of this information to merchants. However, if you couple JTBD theory with machine learning, IoT, and big data, the systems may determine the customer’s JTBD without revealing the customer’s private data to other humans.
Humans currently overshare in social media; they allow systems to capture their data to give them discounts and better deals – it’s not so far fetched to develop a system which, after gathering a large amount of data on a customer’s explicit and implicit desires, design a perfect product for the customer which solves the customers job-to-be-done. Using AI to determine a customer’s JTBD may be more accurate than having humans do it since these humans hold biases.
For example, a customer wishing to own a home (a JTBD) isn’t interested in applying for a mortgage, so an “automated pre-qualification based on a customer looking for a home, gleaned from them searching Trulia and visiting Open Houses” may be a good product for a bank to provide. Or a retailer providing a service that automatically records a customer’s product usage and seamlessly delivers a new product when the product is consumed, at the exact time, place, quality, and price the customer desires, all without human intervention into the customer’s private data?
Combining the JTBD theory, machine learning, ambient sensors, and big data may finally create an autonomous “jobs” engine, seamlessly delivering customers’ desires. Who would not want to live in a world like that?
