Are you going to keep making strategic investments based on guesses?
There are two major worlds in innovation...
The old world
The new world
The New World
There are two camps in the new world:
The guessers
The knowers
The Guessers
The source of your insights is very important. The guessers value their own assumptions and opinions. The data they collect is stored on sticky notes which are collected in a workshop from a group of people that do not represent the market in any way whatsoever.
Do I really need to elaborate on this? The rigor is theater, and the results have not shown a demonstrative, innovative, or competitive advantage. In fact, product and service failure rates are still as high as ever.
But, it feels good.
The Knowers
These folks are typically not designers or product managers. They are front-end of innovation (FEI) focused. This is a a huge distinction. No offense. They don't operate with solution-space constraints at all.
This group is also much smaller for a couple of reasons. The first is that these people are simply wired differently. The second is that the secrets have been held close to the vest for decades.
For the most part, there is far less theater involved because the actual customers and their actual struggles are showcased in granular and concrete data, and this is done at an appropriate scale.
The key difference is in the source of the data - although the modeling of problems is also significantly different. Within this group of knowers, there are different camps who are experimenting in different ways; based on the feedback and observations in their studies. This is to make the approach even better, and hopefully simpler.
The Power of Real Data
Many of you have seen the dots-on-a-plot marketing imagery for Jobs-to-be-Done research. They are very enticing; but not so much actionable. There is a lot more going on behind the scenes that you need to understand in order to make proper decisions, like:
Q: “How large is the opportunity?”
Below, you can calculate the size of the market quite simply. The people willing to pay at least a month's pay is 24.0% + 37.9% = 61.9%. This is pretty amazing given that the company who sponsored this work is in an industry that is considered to be mature (actually at 0% CAGR).
How could this possibly be?
Could we have figured this out in a workshop?
How about with a fancier canvas?
Some other important questions
Q: "What are the top 'N' things (out of hundreds) I can address today to add value?"
This is an important question that cannot be answered with an algorithm. The plots we typically look at can have metrics that have the same score while being on completely different parts of the landscape. This typically denotes a different strategy. So, what do you do in this situation?
Well, an experienced data scientist would tell that you need to use an ordinal model with ordinal data. Ranking is a great way to approach this question since your stakeholder just wants to know what the top things are that they should address.
The question is how to rank them. We have a simple method, although it's not as simple as using RANKX().
Again, if you workshop this, you're just guessing. And if you use a metric model grounded in an algorithm that discards data, and over-indexes other data before making any calculations, you're at risk of over- or under-stating opportunities for improvement (in either direction). Sometimes, significantly.
So, there are criticisms for both camps. Fortunately, there are those of us who learn from these mistakes, and others that just double-down for whatever reason. We test, iterate, and set new standards continually.
Q: "What are the top N things I can address today that remove cost and complexity?"
This doesn't necessarily mean that you need to create a disruptive innovation. It simply means you've listened to too many lead users who do not represent the vast majority of your market. You're definitely at risk of being disrupted, though.
If you want to pull more non-consumers of your solution into the brand, you should be looking for the top measures that are being overserved, eliminate them, and use the money you save to reallocate it to value-adding priorities.
Simple...
...unless you only have sticky notes.
Q: "Are there groups that struggle differently, and how do I describe them?"
FYI - good segmentation almost never looks like the plot you see above. That's marketing material 🤣. If you have a segment where you need to add value to all of your success metrics, good luck! It's better to see overlapping segments in order to add value to a larger portion of the total market.
Segments are difficult to understand if we just look at success measures and dots-on-a-plot alone.
Who are these segments?
Can we develop a persona of sorts to describe them in layman terms.
Above, you'll see a heat map for situational factors. The color patterns (this isn't a good example) will point you to those factors with the most (or least) impact on a needs-based segment.
We capture and prioritize situational (and other) factors on a scale of impact so we can heat map them to understand how they influence the ratings of success measures in different groups. We also do statistical analyses, but this simple approach is very easy for stakeholders to understand and can point to factors that are more causal quickly, and visually.
This is where your data-driven personas will come from.
Data Flexibility
If planned properly from the beginning, your resulting data model can be so much more powerful than a board of sticky notes. We can look beyond needs-based segmentation as well; by plotting our data using any dimension that we've included in the survey.
You never know what you're going to find. The data in the included imagery is based on a very mature market and is also an incomplete sample.
There are dozens more dashboards designed to answer specific questions defined in an analysis plan (follow link if you'd like an AI accelerator for developing analysis plans). No sense boring you with those details since many of you love your workshops and probably didn't get this far 😉.
If you'd like to learn more about this approach to Jobs-to-be-Done you can do any of the following things:
Subscribe to my other newsletter where I go deep into the how-to’s. It's called Zero Pivot Problem Solving.
You can sign up for my Masterclass: Eliminating Jobs-to-be-Done Interviews with Artificial Intelligence. Any students that enroll will get free access to my new course called Zero Pivot Problem Solving (sign up for the wait list if you prefer), which will include updated AI methods as well as a complete how-to guide that even goes into the stuff I just wrote about in this article in great detail (the survey construction, data-modeling, and analytics). No one else will do that! There might even be some Power BI formulas or template reports 😉. But only if enough people express interest. Otherwise, why bother? 🤷🏻
If you're more interested in customer journeys, I've created an accelerator with universal models for 17 distinct customer journeys. You can access that here. Note: this is being updated to include a downloadable PDF and a completely regenerated (and improved) set of value models. Coming soon, stay tuned.
Finally, feel free to reach out to me directly. I'm happy to spend 30 minutes with you to talk about your challenges. Maybe I can help. Here's my calendar: https://jtbd.one/book-mike
The more people that can learn how to do this - or, learn how to bypass the process and move on to more valuable endeavors - the better off we'll all be. We currently waste a lot capital on failing, and should be holding that capital in reserve for more well-deserved investments.
Best,
Mike