The Art of Interpretation: Using pattern recognition to identify demand spaces.
The next big enhancement to the MotivBase trends platform Nothing beats having a client of yours move to a new company and hire you almost instantly. It speaks volumes about the impact of our work. But sometimes, when a very senior person moves into a new role, they find themselves jumping on multiple innovation trains that have left the proverbial station. Take 6 months ago. One of our favorite SVP clients gets hired as the new Chief Insight Officer for a restaurant company that shall remain nameless. They had spent almost 14 months developing a new plant based product but our client looks at what they are about to go to market with and hits pause. Something just isn’t sitting right. One of our favorite SVP clients gets hired as the new Chief Insight Officer for a restaurant company that shall remain nameless. They had spent almost 14 months developing a new plant based product but our client looks at what they are about to go to market with and hits pause. Something just isn’t sitting right. He brings us in. We spend 5 days decoding the dominant demand spaces, looking at what is really driving consumers to embrace plant based solutions, and look to see if both the consumer and their needs ladder back to what the restaurant can deliver. It’s like oil and water. They just don’t fit. Our client has been on the job for less than a month and he sticks a pin in their biggest product launch in 3 years. As you can imagine, we were very popular with the legacy team. When we asked our client how he felt about having to kill the launch so late in the process, he said “The only feeling that matters is do I feel uncomfortable, and am I making sure that we never settle. That’s what innovation is all about.” I’ll tell you what happened after the launch in a minute, but this statement has always stuck with us. In fact, as we have looked at our own MotivBase Trends product, we made a promise that we would never settle. A while back we wrote a post about our company’s focus on using AI and Big Data to continuously improve the quality of observational ethnographic insight and minimize researcher interpretation biases along the way. In that we explained how we’re using machine interpretation in the right way, so as to force the researcher to follow the consumer’s breadcrumbs on any topic or trend being examined. Today we’re excited to announce MotivAI - a feature that allows our technology to conduct another layer of interpretation and further help the researcher or user understand the topic or trend under study and get to meaningful outcomes even faster. MotivAI is an experimental feature, and it will continue to evolve over the coming year. But this first version is already incredible because it quickly conducts a pattern recognition analysis and points out demand spaces that would take a user a bit more time to do on their own. Furthermore, even in cases where it doesn’t quite get the clustering right, it still speeds up the process of identifying clusters and demand spaces for the user by providing the added context needed for the user to quickly grasp what is being talked about and why. Let us look at a quick example - If we run a search for the topic of probiotics - The topic universe as usual tells us about the contextually relevant meanings that consumers inadvertently associate with the topic of “probiotics”. Typically, as a user, you’d spend some time analyzing this topic universe, to identify patterns or clusters of topics that indicate a broader set of meanings. You’d then run a bit more of an analysis and note down some conclusions. It’s not a tedious process, but it does take some time. To be clear, we’re not trying to take away from this important step - it’s critical in cases where a really rich deep dive is warranted. But in cases where a user is looking to get a quick peek at the major demand spaces, this step adds time that our users may not always have at their disposal. This is the problem MotivAI attempts to solve (and does a pretty good job of it already). So let’s look at what it does. If you click on the newly visible AI option, our AI Anthropologist gets to work and takes about 2-5 minutes to run a clustering analysis to identify clusters of meanings that arise out of the topic universe. In the case of the topic of probiotics - here’s what it identifies in about 2 minutes. It plots a series of machine-generated clusters into a chart that allows us to quickly see those sets of meanings that are experiencing growth versus those that aren’t. Let’s explore one such cluster in detail here - anxiety. On the top left, the machine provides a level of interpretation to help the user understand the emerging theme/meaning around anxiety. On the bottom right it provides detail on the specific contextual topics it examined in order to arrive at the interpretation. MotivAI also visualizes these topics at the top right so as to provide a quick visual sense of how the different topics spread out on the maturity curve. A further glance at the bottom right quadrant further validates the power of MotivAI. In this example, it has identified clusters around skin health, mental health, supplements and nutrition, sugar intake, and even gluten tolerance. All incredibly powerful interpretations considering it’s machine led. Of course, not every search you run will yield such obvious results. MotivAI still has some limitations and this is only version 1. But we’re proud of how far we’ve already come on this journey to continue to make observational ethnography at scale more accessible and more agile than ever before. If you’re a client, this feature is already enabled. We are excited for you to try it and take it for a spin. Just remember to be patient and give MotivAI 2-4 minutes to run its analysis. So… what happened with that restaurant we mentioned earlier? Well, their biggest competitor went to market with a very similar plant based product. At first, the team was up in arms, thinking they had missed a huge opportunity. Until their first quarter sales reports came in. The product was pulled from the menu in less than a year.