Location, Location, Location
- 11 May, 2001 13:46
At the progressive Corp., CIO Ray Voelker has run a pilot auto insurance project that studies drivers' behavior here and now, rather than waiting to catch up with traffic tickets and accident reports. In the Autograph project, Progressive places a wireless GPS device in the dashboards of 500 cars of policyholders in Texas. By tracking where those cars go and when - on the highway in the morning, late at night in a bar-heavy area of town, wherever - the company determines the insurer's risk and the car owner's premiums.
Progressive, based in Mayfield Village, Ohio, is among a growing number of companies looking to harvest business value from layering location information onto applications of wireless technology. Thanks in part to federal regulations due out this year for the mobile phone industry, these companies are betting that location will do for wireless what the Web browser did for the Internet. IDC (a sister company to CIO's publisher, CXO Media) estimates that the market for location-based services - now at US$600 million - will come close to US$5 billion during the next three years.
A potential boom for wireless location technology has been on tap since 1996, when the Federal Communication Commission (FCC) mandated that 911 emergency calls from cell phones be traced as landline phones can be. The reason is clear enough: Getting medical or police help to the right place in a hurry is a good thing, of course. With the deadlines for the FCC mandates coming due, an array of companies is exploring what else they can do with location information. Early answers range in scope from the local (buzzing a smart-phone-toting consumer to watch for the coffee bar coming up at the next corner) to the global (alerting a ship's crew that it needs to sail to port for urgent engine repairs). Not surprising, privacy advocates voice concerns along with cheering the prospects of tracking children and Alzheimer's patients.
According to Maribel Dolinov, a senior analyst at Forrester Research, the next few years will bring ubiquitous location information that works with everyday devices like cell phones and PDAs - at little or no cost to the consumer. But while the prospects look great and the concept seems simple (injecting handheld or embedded wireless devices with geographic data), we're still in the early phase here.
Context is EVERYTHING
"All location is," says Iain Gillott, president of Austin, Texas-based iGillottResearch, "is a piece of data that is entered into an application and increases its value." But since most companies are still familiarising themselves with rudimentary wireless applications, incorporating location information is still at least a year and a half away, according to Gillott. There's another level too: This expected learning curve that Gillott cites for user companies doesn't include all the work that telecommunications carriers like Verizon, AT&T, Sprint and others must do to meet the FCC's requirements - and even then limitations on their quality of service could trip up location trackers.
The trick for the forward-thinking CIO is to learn about the new technology and start thinking of the ways that it could apply to his or her business - before competitors do.
This kind of thinking drives Progressive, the US$6.8 billion insurer. Voelker says that insurers classically use factors such as motorists' safety records, commuting distances and age as proxies for calculating the risk of an accident. If widely implemented, the GPS devices could give his company an edge over rivals. "Autograph comes closer than anything else about predicting how likely [motorists] are to get into an accident. And if they don't have a high likelihood then they shouldn't pay a lot for a premium, and that's a good thing. If we are the only ones who can figure out how to give the good drivers a price break, then that's good for us and good for the drivers," Voelker says.
Good, but not free of system costs and technical challenges. Progressive outsourced the work of satellite-based location identification to Irving, Texas-based ATX Technologies. The Autograph pilot, which ran from August 1998 to April 2000, showed that most drivers' insurance rates decreased. But Voelker says that Progressive, which operates nationwide, isn't ready to talk about a broader rollout of the Autograph system. It turns out that building the system to handle the new location information was the hardest part of the project. Since the Autograph project requires calculating rates in a fundamentally different way from traditional rates, it would have required a costly retrofitting of the legacy billing and policy processing systems. Voelker says Progressive weathered the pilot with a new system that could handle up to a thousand policies. But a large-scale implementation of Autograph would call for a fresh (and sizeable) IT investment.
The wireless location services that are available today are delivered through private contractors and are satellite-based. Location services, like those piloted by Calgary, Alberta-based Cell-Loc, dreamed about by marketers and championed by the FCC's Wireless Telecommunications Bureau, are at least a year and one technology generation away. Still, there is preparation that a CIO can do today to help his or her organisation take advantage of location technology when it does become widely available.
First and foremost, start making existing applications ready to go wireless. "You have to be doing something wireless to care about location," says Forrester's Dolinov. "Location isn't different; it's just the next generation of wireless services. It is an evolution. You start out with wireless services and you layer location-based services on top of that." Even location pioneers such as Progressive have followed this advice. The insurance giant has exchanged claim information and accident reports wirelessly with its fleet of about 1,500 customer-response vehicles since 1994.
Also check with telecommunications providers and wireless application builders to see what their location plans are. Dolinov suggests having systems integrators that specialise in mobile technology describe their strategy for location-based services. Ask "when they think they are going to have location services available and who they are working with. Make sure you know their time frame so that you are not left behind when that functionality is available," she adds.
You may find that you gain experience during the wait for important functionality. Kozmo.com, the New York City-based delivery service, loves the added capabilities it gets from being able to communicate wirelessly with its couriers in the streets, says Roger Smith, vice president of technology. Kozmo.com, which promises one-hour delivery, outfitted its messengers with wireless devices in July 2000. "Without any foresight into [our customers'] random demand it is a challenge to respond to it. And so these tools allow us to be much more nimble and retask our people so that they don't have downtime," Smith says. Kozmo wants to take the dispatching tool a step further, adding location information to its logistics planning software as soon as a feasible option becomes available in New York City where tall buildings render GPS unusable.
While location information from major vendors isn't yet available and probably won't be for some time, a handful of location startups have taken matters into their own hands. Cell-Loc, for example, built its own network in Calgary. One application was a consumer product that gave users directions to nearby restaurants and other requested information. Cell-Loc has also beta-tested location-based applications with fleets of trucks at companies local to the Canadian city and will soon begin a U.S. pilot in Austin (Cell-Loc officials won't release the names of their pilot companies). Application vendors such as ATX Technologies have also contracted with location technology vendors to incorporate that data into applications, which they in turn will sell to customers.
Voelker, the CIO at Progressive who worked with ATX, says that understanding the different parties involved and exactly what they do is crucial if you are going to depend on location technology. "CIO's worry about end to end, from the time somebody hits a button to the time I get the response I want," he says. "But the wireless providers talk about what their reliability and response is. The local [carrier] guys will talk about coverage in their area and on the device, but all you care about is end-to-end response time. And it can be challenging to get those things all lined up."
Voelker says the best way to meet this challenge is to bring experienced telecom people into the contract negotiations, plan for failed transmissions, and stay current with the latest wireless middleware. Most important: Leave room to change vendors so that you're not stuck with a company that's lost when it comes to location.
As Cell-Loc's experiments with trucking companies show, location technology has clear implications for companies with heavy logistics investments. Sensing this, Rolls Royce Marine, a multinational unit of the British engine maker, has designed a system that uses location technology to provide preventative maintenance to ships at sea. "We probably wouldn't put a big effort into just ship location," says the company's Norway-based CIO, Geir Balsnes. "We need to have something on top of it." The simplest benefit is better knowledge of when a ship will come into port. Typically, a ship notifies the maintenance staff at Rolls Royce Marine once it arrives. The staff then mobilizes and services the vessel. With location information, Rolls Royce knows when a ship is about to enter port and can have its service people ready, reducing maintenance time and saving everyone money.
Ship to SHORE
But that's just the beginning of Balsnes' vision. Adding location information to Rolls Royce's proprietary predictive maintenance software, which monitors the condition of a ship's parts, would allow for unprecedented levels of service. "Instead of having a breakdown," says Balsnes, "you could see that a propeller shaft might break or that an engine might stop." By combining this insight with detailed location information, "you can guide them to port before they have an accident at sea."
As with all applications using wireless technology, limitations such as spotty coverage in certain rural locations remain. According to Gillott, there aren't enough base stations in rural areas to get an accurate network-based location readout. And GPS doesn't work well indoors or in cities with big, shadow-casting buildings. "If you're going to get mugged," he says, "don't get mugged indoors or in New York City."
And all the IT executives interviewed for this article returned to the idea that location is another feature, not an answer by itself. Even when experimenting with wireless technology, evaluate how location-based services would or would not help your business strategy. "I know where I am right now," says Gillott. "I don't need my wireless network to tell me." But with a little imagination, there's no telling where companies using location information will find themselves.