Power struggle Customers, companies, and the Internet of Things


As the Internet of Things (IoT) permeates people’s daily lives, potentially useful information can now be created at every turn. But sometimes customers, companies, or both can find themselves disadvantaged by IoT-enabled deployments. Host Tanya Ott spoke to Michael Raynor and Brenna Sniderman of Deloitte Services LP on how to balance this power struggle.

The Press Room
Subscribe on iTunes
Listen to more podcasts

Learn more about the Internet of Things
Read the full article
Subscribe to receive IoT insights


TANYA OTT: This is The Press Room, a Deloitte University Press podcast on the issues and ideas that matter to your business today. I’m Tanya Ott, and today we’re talking about the Internet of Things.

MICHAEL RAYNOR: Your car, your house, your sneakers, your tennis racket . . .

BRENNA SNIDERMAN: Customers know that their data is being gathered and analyzed, but they don’t want to feel watched. And they don’t want to know that they don’t have a say in the matter.

MICHAEL RAYNOR: Those disclaimers that some people tend to just walk past as a necessary check box before they can get on to what they really want to do—I think it will be increasingly important for people to understand precisely what it is that they are agreeing to.

TANYA OTT: The Internet of Things: What is it? How can consumers protect their private data . . . and how companies can make sure it’s a win-win proposition for them and their customers?

TANYA OTT: I’m a customer. Every morning I walk, and I wear a fitness tracker. I don’t usually think about where all of that information my wristband collects ends up. But I’ve been thinking about it a lot since I recently talked with Michael Raynor about the Internet of Things. He’s director of research at Deloitte Services LP and author of books with titles like The Strategy Paradox and The Innovator’s Manifesto.

MICHAEL RAYNOR: The Internet to date has really been an Internet of people. The “things,” if you will, at the end of the Internet that are inputting the data and consuming the data have been, well, human beings. And we’re actually really good at dealing with messy, fuzzy, difficult-to-work-with data, and that’s been a good thing because it’s allowed the Internet to really take off and find ways to refashion industries, create entirely new opportunities for companies to serve customers, and for people to lead interesting lives in many different ways. But the Internet of Things puts objects, inanimate objects, as data creators and data consumers. And really that’s only been possible over the last few of years as a result of improvements in a whole suite of technologies that have put sensors on objects.

TANYA OTT: Everything from my fitness tracker to your car. Did you know the average car has about 200 sensors in it?

BRENNA SNIDERMAN: The connected navigation systems would be one example where they can spit out, based on information of where you are, and other automobiles might be able to tell you real-time traffic, where to go. And as you’re getting closer to your home, alert your home to start warming things up . . .

TANYA OTT: That’s Brenna Sniderman, senior manager at Deloitte’s Center for Integrated Research. She and Michael Raynor write about these sensors and the ways companies use the data they collect in a new article titled “Power struggle: Customers, companies, and the Internet of Things.” And there is a struggle, because as computing gets simultaneously less expensive and more sophisticated, companies can now aggregate and analyze data in ways that actually allow machines to make decision and take actions. Take the “connected home” example with a “smart thermostat.”

MICHAEL RAYNOR: It would decide when it’s going to turn up or turn down the heat or air-conditioning in your house based on your observed patterns of behavior and the price of electricity in the moment, such that the algorithms could come to a conclusion that, in fact, it’s okay to let the temperature go up a couple of degrees, because frankly it knows that—at the risk of anthropomorphizing your thermostat—you actually won’t mind that much, and that the marginal increase in discomfort is more than offset by the reduction in your utility bill. So it’s not just giving you information so that you can make decisions. It’s actually coming to the point where these things appear to make decisions themselves.

TANYA OTT: Kind of reminds me of a super cheesy made-for-TV horror movie I saw as a kid in the 1970s. A “house of the future,” outfitted with all kinds of sensors, starts turning on—and killing—visitors and even the family who lives there.

MICHAEL RAYNOR: It’s fodder for all form of science fiction, and the good news is there’s no reason to think that anything out there is sufficiently malicious to make that a possibility. (laughs)

TANYA OTT: However, you know people are really worried about all that personal data that’s being collected. What are companies tracking about me? What are they doing with all of that information? Do they have a legitimate concern?

BRENNA SNIDERMAN: You know, that’s something we talk about a lot: the notion that companies can use consumer information to capture value at the consumers’ expense. But if they go too far, they can tip the balance too far in their own favor, which can create a backlash among customers. But on the flip side, if they’re too cautious and they don’t gather enough information, then they’re not really benefitting from using IoT.

MICHAEL RAYNOR: What we’ve observed is that, in some instances, consumers may actually find themselves with the upper hand, and, on the face of it, you might think, that’s great, power to the people. Except for the fact that if there is an imbalance in either direction, that actually undermines the long-term viability of the market because one half of that exchange is simply going to opt out, right? They’re simply going to say, there’s not enough in this for me. That’s something consumers might conclude, and some markets will break down because people will cease to share their information. Or companies may conclude there’s not enough in it for them, and so the benefits that might otherwise be provided would evaporate.

TANYA OTT: So how to find a happy medium? One place to start might be with a grid Raynor and Sniderman developed. Picture this: four boxes, 2×2. The X-axis is the value to the customer; the value is greater as you move to the right. The Y-axis is the value to the company; it gets greater as you move up.

BRENNA SNIDERMAN: The goal is for both the customer and the company to capture a high amount of value.

TANYA OTT: Where you don’t want to be is in the lower left corner of the grid—the box they label “Gridlock.” And surprisingly, the real-world example they give is that super-smart—or super-creepy, if you saw that horror flick—”smart home.”

TANYA OTT: So let’s break those down. The one that has the lowest value to both company and customer is what you’re calling “Gridlock.” And you use an example: smart home. Why is a smart home not value for either?

BRENNA SNIDERMAN: That’s something that we really struggled with, because I think the smart home is something that could have a tremendous amount of value for companies and customers. The customer can save energy. It’s so much more convenient. You can do things like, as I mentioned earlier, as you’re driving home, your garage door opens, your security system turns off, your AC turns on for you. Even something as simple as when you’re away from home on vacation, and it gets below freezing, you can make sure that your pipes are dripping water so they don’t freeze. And an energy company would be able to understand who’s using what and where, so they can better allocate resources and plan for power surges. The makers of smart home devices can start to get a better understanding of their customers’ behaviors and serve them better.

TANYA OTT: Warning: Here comes the “but.”

BRENNA SNIDERMAN: You have the replacement challenge: The refrigerator or the thermostat, you’re replacing them probably every 20 to 30 years. So the period of adoption of IoT for a refrigerator . . . I mean, if I just got a new refrigerator five years ago, I’m not going to go out and buy a new IoT–enabled fridge. What a lot of connected technology companies are doing now is creating devices to enable you to retrofit devices in your home to make them IOT enabled, so that’s one challenge that is being met. The other challenge, however, is the challenge of interoperability standards.

TANYA OTT: Basically, right now, because it’s so new and there’s no real standardization, many IoT devices can’t talk to each other. Your car may be smart. Your thermostat may be smart. But if they don’t speak the same language . . . well, you can’t take advantage of all those smarts. Sniderman says that’s something the marketplace—or regulators—might shake out down the road.

TANYA OTT: Moving along the spectrum of the grid, there’s the box labeled “Customer is king.” Sounds like a good thing, right? But remember what Michael Raynor said just a few minutes ago?

MICHAEL RAYNOR: If there is an imbalance in either direction, that actually undermines the long-term viability of the market . . .

TANYA OTT: Brenna Sniderman says a good example of this is “connected retail,” where retailers use customers’ browsing and buying behaviors on apps, websites, and in brick-and-mortar stores to customize the shopping experience for them.

Forrester Research predicts that by 2016, connected retail will influence 44 percent of retail sales. The Guardian newspaper reports, “It has the potential to revolutionize the industry and transform how retailers operate and how they connect with consumers,” by improving operations, streamlining processes, and leading to greater efficiencies.

BRENNA SNIDERMAN: Because if you have a lot more information about the customer, you can provide them with more personally relevant information: A customer walking into a store, and—based on browsing and purchasing behavior they’ve done on the app and the website, other purchases they’ve done in the past, how they’ve reacted to offers, and personal information they’ve provided during past interactions—the company can aggregate that information and tailor offers specifically for that person. And they can create a 360-degree for that customer, help guide them through that physical store, provide them with exclusive offers, and so on and so forth.

TANYA OTT: Sounds great, right? Okay—wait for it—here’s the “but.”

BRENNA SNIDERMAN: At the end of the day, the retailer is still more or less selling the same products and services. They’re using the Internet of Things as another way to make it more convenient for the customer to shop with them, to purchase their products and services, but, at the end of the day, the customer can still go to another retailer. There’s nothing really keeping them with that particular retailer. So the customer still has all the power here. It still depends on a customer making a conscious choice or a conscious action.

TANYA OTT: So, by Raynor and Sniderman’s estimation, it’s high value to the customer, but lower value to the company. On the other end of the grid spectrum, there’s a box they label “Hobson’s choice.” Remember that? There’s a famous example.

BRENNA SNIDERMAN: Henry Ford with the Model T, where they have that famous statement, “A customer can choose any color for the car they want, so long as that color is black.”

TANYA OTT: The customer has the appearance of choice, but there’s really only one choice they can make. The example they give is car insurance.

BRENNA SNIDERMAN: You have a choice in as much as you can choose between insurers, but, at the end of the day, you really don’t have a choice because you have to have automotive insurance. That’s the law.

TANYA OTT: Remember those 200-plus sensors embedded in your car?

BRENNA SNIDERMAN: Many automotive insurance companies are starting to use what’s called “usage-based insurance,” or UBI, where there’s a device in your car. Increasingly it’s either in your car when it rolls off the showroom floor, or the company will mail you a device to place in your car when you purchase insurance with them. That will track how you drive. So obviously this creates an issue for drivers. Personally I would recoil at wanting to be monitored. You know, I think I’m a great driver. Everyone thinks they’re a great driver. (laughs) You don’t want to find out that compared to everyone else you’re actually a terrible driver, and we’re going to have to charge you more! I think that’s part of the challenge.

MICHAEL RAYNOR: There’s two sides to this conversation—and that’s how insurance works now. Insurance companies use all manner of proxies in an attempt to understand the risk you may or may not be. So people with severely blemished driving records as a result of a history of at-fault accidents will find themselves paying higher insurance premiums than people who have a spotless record. And most of us have an intuition that that seems perfectly reasonable—that a riskier driver should pay a higher premium because there’s a higher likelihood that they’ll be making a claim. Fair enough.

BRENNA SNIDERMAN: Once [insurers] gather data of sufficient scale across enough drivers and sufficient scope across enough tidbits of information, they can create a much, much, much more accurate understanding of risk. For example, if everybody is running red lights, and you’re running red lights too, your risk is the same as everybody else’s. But if nobody is running red lights, and you’re running red lights, then you have a higher risk profile than others. Once they understand your behavior relative to the larger pool of drivers, they can assign you a risk profile and thus adjust your rates.

TANYA OTT: But, Raynor says, if an insurance company collects too much information about you . . .

MICHAEL RAYNOR: If people find these data being collected on a near-continuous basis on all of their driving behavior, precisely what will that be used for? And does it respect our notion of fairness in terms of what we think is reasonable and speaks to the responsibility that an individual might bear versus a somewhat soulless statistical algorithm saying this is what you’re going to pay as a result of what we think the algorithm might suggest?

TANYA OTT: So where is the sweet spot? It’s in the top right quadrant of the grid. The data collection results in high value for the customer and the company. Their example here also comes from the auto industry.

BRENNA SNIDERMAN: We just talked about insurance and what a challenge that can be, but another challenge for me personally is car maintenance. I always, always, always have the experience of bringing my car to the shop or to the mechanic for regular maintenance, and it always turns out there’s something else wrong that I had no idea about. Not only that, but I don’t actually know if it’s an issue or if they’re just saying it’s an issue because they’ve got my car up on blocks, and they can do whatever they want with it. And I have honestly would have no idea either way.

We talk about the idea of vehicle diagnostics, where they could allow a car to self-identify surface issues rather than rely on me as a customer saying “I hear squeaking, I don’t know what it means,” or a mechanic says, “Yes, you definitely need everything replaced in your car.” So it creates a level of objectivity where the customer can really understand what’s actually wrong with their car and know when they need to take it in to get service. That’s a positive for the customer, because they know what’s wrong, and they can go to the mechanic or the maintenance guy and articulate what’s wrong. And they know when they need to bring it because they will get an alert. We talk about how it might be taken a step further by potentially syncing with an owner’s calendar on their phone to schedule a dealership appointment when they know it’s going to be a convenient time for them. And then scheduling a loaner vehicle so they really don’t have to do anything but show up, drop the car off, and take a new car that they can use for the time being.

Where the positive side comes in for the company or for the auto dealership is that it creates more touchpoints where they can see the customer. The customer comes in more regularly for their maintenance. It can build more of a rapport with the customer and build more customer loyalty to ideally have them go to [the dealer] rather than go elsewhere for a tune-up. [The auto company] can then use the data regarding vehicle maintenance issues and identify if they’ve gathered data with enough scale and scope: whether there are systematic malfunctions across a larger pool of this similar automobile that are worthy of large attention. Does every single car of this make and model have to get its brake pads replaced at this particular time? That’s something we need to look into. So the benefits for both sides are pretty positive.

TANYA OTT: Let me give you a hypothetical [situation] because we talked a lot earlier about fitness trackers. The fitness tracker stood out to me because it is in some ways similar to the car insurance example that you have, in that if I’m being tracked all the time and my health insurance provider is looking at, did she do 10,000 steps today and the next day and the next day, and then setting my rates on that, I think that’s something that consumers could be pretty upset about!

MICHAEL RAYNOR: I can see why that’s the sort of thing that would occur to you and others, for that matter. When you’re thinking about the data that you gather . . . what that reveals about each of us and parts of our lives that we wish to keep private for entirely other reasons, compared to what some would see [as] the simply logical extension of the way a lot of this insurance is priced in the first place. On the life insurance forms, it asks you, do you smoke, or don’t you? It asks you, do you drink? If you drink, how much alcohol do you consume? So they ask you these questions already. Where we run into a problem is at what point does the quantity and the quality of the information that we’re generating lead us to a fundamentally different set of circumstances? Rather than simply collecting what would otherwise seem to be reasonable data in order to accurately price a product, we’re back to this whole notion of an exchange of value. After all, if insurance companies can’t price the insurance correctly, then naturally the insurance will be unavailable. And at what point has it simply gone so far that now it feels like a fundamentally different problem, and it bumps into some of the issues that you mentioned?

TANYA OTT: Sometimes that point comes when companies move from responsive analytics—for instance, sending you a push notification on your smartphone about a sale on bread one aisle over in the grocery store—to predictive analytics.

BRENNA SNIDERMAN: Which can sometimes, to a lot of people, to some people, feel a little bit creepy, and this is where we get into the overstepping. Customers know that their data is being gathered and analyzed, but they don’t want to feel watched. And they don’t want to know that they don’t have a say in the matter. If a company is predicting what they think you’re going to do and sort of staying one step ahead of you, it can feel a little bit strange and discomforting to people.

MICHAEL RAYNOR: I think that ultimately pretty much everything that’s important to us will be touched by these technologies—that as sensors get smaller and smaller and require ever less power in order to generate data, we will find uses for data generated from things that we probably don’t even imagine today. That always seems a rather safe bet, that human creativity generally finds a way to apply new technology to just about anything of significance, and I think this is no exception.

TANYA OTT: So, buyer beware, caveat emptor and all that! Michael Raynor and Brenna Sniderman offer more for companies thinking about getting into the Internet of Things in their article “Power struggle: Customers, companies, and the internet of things.”

I’m Tanya Ott for the Press Room, a production of Deloitte University Press. We post new podcasts twice a month, and if you subscribe, you won’t miss a single one. You can check out our archive at dupress.com where we recently asked the question: What happens if we run out of water?

Please tell your friends and colleagues about the podcast—and tell us what you think about the issues we’re covering. You can leave a comment on the podcast page, or email us at podcasts@dupress.com. And if you’re on Twitter, you can find us @du_press. We’d love to hear from you . . . really, we would! Thanks for listening, and catch ya next time.

This podcast is provided by Deloitte LLP and its subsidiaries and is intended to provide general information only. This podcast is not intended to constitute professional advice or services of any kind. For additional information about Deloitte LLP and its subsidiaries, go to Deloitte.com/about.