In the age of Internet-connected devices, data-gathering capabilities have never been more abundant, and in the age of big data there is a premium being placed on technology that can make sense of the data that’s available and to delight users. Mike Vladimer, who co-founded the Internet of Things-focused IoT Studio at Orange Silicon Valley, loves this problem and has developed a framework for extracting real value from IoT data.
Mike will be speaking about his framework at the San Francisco Internet of Things Meetup on May 31. Ahead of the talk, he offered some insights about how why he created the framework and how he understands the problems that it tries to solve.
Orange Silicon Valley: What problem does your framework for using data from the Internet of Things solve?
Mike Vladimer: The problem with the Internet of Things is there are a lot of IoT devices that stink. There’s a lot of people that say the Internet of Things is useless, it was abuzz and it’s over — there’s a lot of people that are disappointed.
OSV: So where does the problem begin?
MV: There’s a problem in the market as evidenced by tweets and articles that we can measure. The second level of that is, “OK, what are people complaining about?” What I would say is there has been a historical challenge where companies would say, “Oh, there’s this really important problem I need to solve. If only I had the technology to solve it. I would love to be able to travel faster, but all I have is this horse.”
For a really long time we’ve been living in this world of horses, and we just didn’t have the technology to do something better, and in that case fossil fuels really unleashed the ability to move faster and gave people access to power. Similarly, we live in 2018, and the challenge has shifted. It used to be, “I wish I could solve these problems.” Now it’s, “I can pretty much solve whatever problem I want to solve. My challenge is to find a valuable problem.”
The framework I’m presenting next week at the San Francisco IoT Meetup at Canopy in North Beach is to solve this issue: How do we find compelling problems? What makes a problem compelling? What makes an IoT solution a compelling solution? Really, it’s a matching challenge.
OSV: Why is the data a problem? What is the problem people encounter when they try to make use of IoT data?
MV: That’s a great question. This reminds me of a great Woody Allen joke. At the beginning of “Annie Hall,” there’s these two old women at a hotel resort, and they’re complaining to each other. The first woman says, “Boy, the food at this place is really terrible.” And the second woman answers, “Yeah, I know; and such small portions.” The joke of course is that if something is so bad, why would you want so much of it? And it’s so funny because it’s true.
That’s kind of how I feel about IoT data: “Ugh, the data, it’s so messy and awful. But at least there’s a lot of it.”
The thing about IoT data as opposed to bad food at a resort is if you have a lot of IoT data and you just present it to a user directly, it stinks. It only solves 80 percent of their problem. It doesn’t really solve the problem. And it leaves them with a lot of work to do.
By contrast, because there is so much data, you can actually start to answer specific questions with it. You can give a 100 percent solution; you just have to be deliberate about that. That’s a lot of what this framework is about.
The idea that I like to use as an example is to think of the common thermometer. We grew up with a thermometer on our window. It tells you the temperature outside. And historically people have wanted to know, “What’s the temperature?” You look at the thermometer, and it says, “It’s 65 degrees.” Great.
Now, with the Internet of Things we can track the temperature — right now it’s 65 degrees, an hour ago it was 64 degrees, and here’s what the pattern looked like over the week. Most IoT devices with do that. They’ll give you a time series graph of data over time that is a little better than the status quo of, “Here’s what the mercury says right now: 65 degrees.” But there’s a lot of work left for the human. It’s an 80 percent solution.
A really great IoT device — and this is what my framework gets at — figures out what the fundamental question is and solves it 100 percent. So when someone says, “What’s the temperature outside?” you have to understand that they’re asking that because they never could ask the real question. The real question is, “Should I wear a jacket today?” And yes, we have the data; we can say, “Yes, you should wear a jacket today.” Not only that, we can say, “Yes, Brian. You should wear a jacket today because you tend to run cold, but your wife tends to run hot.”
OSV: Because you’ve applied a personalized model to understanding my data.
MV: Exactly. So step one is identifying the right question. Step two is answering it completely and not just leaving you with, “Here’s the temperature right now; you figure it out.” It’s not a lot more work from the IoT system/device side to get you that answer. But to me — and this is the whole idea of extracting value from IoT — a little more work creates so much more value for people.
OSV: Let me expand on that question. In what kinds of setting or industries do you see better use of IoT data have the biggest impact.
MV: I get that question a lot — more typically as “Where is IoT going to be hot?” So my answer for that is when people ask that question I think they are coming to that from the general reference point of a laptop or smartphone. And I don’t think IoT is great for that. I think IoT is much more specific. For instance, let’s stick with our thermometer example. The thermometer is nice because it’s a general tool; but for you the question might be “Should I wear a jacket?” but for Bob the question he cares about might be “Should I go to the beach today? Is it going to be sunny and beautiful?” It’s the same underlying temperature data. But the analysis answers very different questions.
So, I think your original question was where IoT is going to be the most impactful. I think IoT has the potential to have huge amounts of impact for very small numbers of people. If you’re diabetic and you’re trying to understand your blood glucose, there’s not that many diabetics in the world, but IoT can help them that much more. The key thing in my mind is that IoT drives at that 100 percent solution.
The killer feature, if you asked Steve Jobs back in 2007, for the iPhone, was telephony. The one thing it could do end-to-end with no outside complications from the user was make a phone call. Now, with ten clicks you order an Uber or make a tweet, but it had that kind of killer use case. And I think with IoT that shedding everything that’s not the killer use case is the formula for success.