We are quickly hurtling towards peak smart product penetration. Smart underwear or egg trays, anyone? As you get more and more smart devices into consumers’ homes, you have a greater opportunity (and responsibility) to use that resultant data. At first glance, this sounds like a dream: actual usage data! Goodbye inaccurate self-reporting and journaling. Real-time data. Data without recruitment and incentives! A market researcher’s dream, right?! Many of our clients find the excitement wears off quickly when the time comes to make order of all that chaos and start solving business problems. There is so much data there and it’s growing. All. The. Time.
As a company that works with big datasets regularly, we understand that you can’t just dip your hand into the data-well and come back with a hand full of insights. It may not be clear to your stakeholders, but it requires a lot of work to move beyond surface-level analyses to fully sift through to the depths of your smart product dataset and find those meaningful insights.
At Escalent, we have four steps for success when faced with the task of using smart product data. Let’s walk through these while thinking about those smart underwear, which can track heart rate, temperature, pressure, motion, body fat, and hydration levels.
It may surprise you to hear that the single most time consuming part of the process is getting the data into an organized, analyzable state. During this stage, it is important to define the following:
In the smart underwear example, we would need to flesh out what the underwear can track and how that manifests as database variables, how often data is sent, whether or not we can append registration data to add a demographic overlay, and how to transfer that data, among other considerations.
Fishing expeditions—where someone roots around in the data fishing for golden nuggets of insight—are rarely fruitful. Like any other good research, setting strong objectives and hypotheses will set the course for the entire endeavor. There is generally no shortage of questions and hypotheses in any given organization but the real task is to collect and organize them. A workshop is a great way to obtain stakeholder buy-in, clarify objectives, and generate hypotheses.
In considering smart underwear, potential hypotheses could include:
Even with big data, it’s not too difficult to run surface-level analytics to obtain a surface-level understanding of the data. But the real power comes when we find meaning in the data. We do this by taking otherwise faceless data and creating defined events, which are events that are happening in the world and can be constructed using the available variables. These defined events are typically limited to occurrences that align with your hypotheses and objectives. This can be done through a deliberate interpretation of a single variable or through the combination of multiple variables.
For example, we can understand body fat averages and other descriptives with an easy analysis, but the real work and fun comes when we take it a step further. How do we know when someone is exercising? What combination of variables can tell us that? Is it a sustained increase in heart rate and motion? What has to happen for us to confidentially say someone has worn their underwear for more than 48 hours? Is it as simple as more than 48+ hours of continuous body temperature readings, or is there more to it? Understanding the data and stringing multiple variables together is no small undertaking, but they are vital to any deep analysis.
Now that we have created defined events, we can observe and track. We know what people are doing, but that begs another question—the all-important “Why?”
Even the best connected device database cannot tell you these things, which is why Escalent recommends recruiting users into an Insight Community. Not only will you receive all the standard benefits of a community (such as quantitative and qualitative research, longitudinal findings, and strong ROI) but you can also follow up on defined events to add context.
In going back to our smart underwear example, consider this sample conversation that could take place in an Insight Community between a moderator and a smart underwear user:
Community Moderator: Hi, Jon, it looks like you were rather dehydrated yesterday before you started working out. Any thoughts on why that might be?
Respondent: Yeah, I got too busy at work and forgot to refill my water bottle. I chugged some water a few minutes before I went to the gym, but I could tell that didn’t work.
Community Moderator: Great feedback, thanks! Out of curiosity, how much did you drink and how soon before your workout?
Respondent: It was a half Nalgene, so about 500 ml. It was about 10 minutes before.
This quick interaction in a community would provide us the necessary context to understand why the user worked out when they were dehydrated: they got too busy during the day to think about refilling their water bottle and their attempts to chug water before their workout were too little, too late. And, if we were to learn that this is a common problem, it would present an opportunity for innovation. In the future, for example, the smart underwear app could alert the user when they are dehydrated, letting them know how much and when they need to drink fluids to rehydrate themselves appropriately before their workout.
Although the above conversation is merely an example of the type of quick, insightful interaction that could take place in a community, it is emblematic of the types of activities we regularly see and manage on behalf of our Insight Community clients. And when you think about it, what we are able to do is pretty incredible. We are able to take faceless data streams and add context for deeper, broader understanding, thanks to the willing participation of community members.
The beauty here is that community members are highly engaged and genuinely want to help. The bond they feel with the brand and the moderators makes them feel comfortable enough to share details that provide important context. In our example, it would take a brave crew to share insights about their smart underwear usage and health data, but it’s not a stretch to ask someone why their refrigerator door was open for five minutes straight or why they changed their smart light settings 12 times in an hour.
We believe communities are the single best way to add context to defined events because of the depth of understanding it unlocks and the degree to which your users will be comfortable sharing details in this format. That said, bespoke qualitative and quantitative studies do provide targeted insights. An example using these methods would be to send a survey to a group of users who are often dehydrated or to host a focus group among those who wear their underwear for more than 48 hours (maybe it would be best to do that one online!).
Delving into your connected product dataset may be daunting, but these four steps can get you well on your way to a robust and meaningful analysis. Escalent is here to partner with you from beginning to end, so please reach out if we can be of help to you. Our Consumer & Retail team is at the ready to support your needs, be they smart underwear-related or otherwise.