InsightaaS: Inherent in the appeal of mobility (and to a substantial degree, of cloud-based Big Data analytics as well) is the concept of location-based services: the idea that it is possible to tie physical location together with the vast resources of cyberspace to create entirely new services — and new sources of revenue.
To explore this issue further, we reached out to Phillip Kaszuba, President at Markham, Ontario-based DMTI Spatial. The firm, which was acquired by European mailroom services giant Neopost in October, has been a leader in providing commercial location-based solutions for nearly 20 years. In this edited interview transcript, Kaszuba traced DMTI’s path through the early days of location-based services, and explains how the technology will have a significant impact on business norms and options in major industries like financial services.
InsightaaS: Location-based services is a hot term today — but DMTI has been in this space for nearly 20 years. What initially brought you into a market that didn’t even really exist?
Phillip Kaszuba: The company started when one of DMTI’s founders, John Fisher, was working with the Ontario government. One of the ministers came to him and said, "Could you figure out a way to help me route ambulances more efficiently around the province? Because we can look at a whole history of ambulances going to the wrong place, taking the longest route, all those types of things." And as John describes it, that's what he really had got hooked on what you could do with GIS data and information and how that can have a huge impact.
So he started by helping the Ontario government build out a digital network for routing ambulance, but he quickly realized that this has wider applicability. So 25 years ago he started a company, and it got merged into DMTI. Today, we've got about 350 customers from coast to coast — including Rogers, TELUS, Bell Canada, federal government agencies like RCMP, Department of National Defence, Public Safety, Health and Human Services, Aboriginal Affairs and Northern Development. We've done well in the banking and insurance community. Over time, we've built some great customer relationships, based on providing them with insights through geospatial data.
About five years ago we were working with one of our key customers, Genworth Financial, on a project helping them to increase the level of automation applied to incoming business, so that they don't have to have underwriting touching every file, they don't need to have an adjudication process that could costs hundreds if not thousands of dollars. Genworth realized that the automation was very dependent on location. They asked for systems that would tell them something about a property, about a home, about the proximity to other things — and we started partnering with them, and it led to an initiative that we have focused on for the last 3-5 years. The core focus is, how do we deliver real-time location information to our customers so that they can automate things like adjudication decisions — answering questions like, "do I want this customer, and what's the best way to get to them?"
If we look at Genworth — basically what they do is provide mortgage insurance for big banks. And they get somewhere around two to three seconds in order to provide a reply to a bank to say "yes, we want to go ahead and do that," or, "no, I can't do it," or "I want to do it but I need to go through this adjudication process." And what's interesting about their business model is that if they don't provide the answer in those three seconds, the request automatically rolls over to a competitor. They don’t get a chance to compete for the business tomorrow: it’s "you make a decision and you live with that decision."
So Genworth really helped us evolve what we'll call real-time access to location information, within milliseconds. Today their standard SLA is 50 millisecond response time. We provide location information and then insert it right inside of an existing process in that time. In another account — one of the major banks — we’ve got a massive SAP deployment. That response time is less than 50 milliseconds, and they have accurate location information [linked processes] to let location information be put right into a mortgage at the initial approval process.
Three years ago we built a new SaaS cloud offering on the Microsoft Azure platform, and we also provide our own hosted environment out of Q9. We've now got about 50 customers that are directly integrated into our SaaS offering. With them, we start by making sure that the information they have is accurate, and then we enrich it with additional geospatial information and other information to help them better understand how location might impact a decision.
InsightaaS: Besides Microsoft and Q9, what sorts of partners do you work with?
Kaszuba: Over the last three years we've solidified some pretty interesting OEM relationships, including organizations like TomTom, Navdeq, Nokia, Apple, Google, Garmin — traditional personal navigation companies. Those relationship are a result of our good topological understanding of Canada. But what we've seen is that these partners are really trying to figure out how do they embed that type of information and those types of insights into applications. We've started to see location services start to be an anchor that lets you look at that high velocity data that is coming to you from the mobile society and from all the applications and apps out there. Companies are starting to use location to build models that let them target customers more efficiently.
The view that we've had is if location can give you this new perspective, it can allow you to look at your customers and your market opportunities differently. We work with third party data sources like Environics Analytics to link location to enriched content like a demographic profile or a sociographic profile or a business profile.
As an example — we did a project about a year and a half ago with Union Gas. They were very focused on eco-friendly products and services and wanted to understand how to best introduce these services to their customer base. We worked with them to understand everything we could about location so that they could introduce the right services to that location. We looked at the size of the actual house, at the age of the house, at the gas consumption for that house…we had a whole series of factors that allowed us to target some very specific campaigns using our technology with their customer understanding that really helped them drive some wonderful results on their marketing programs. Last year alone they had a 400% increase on the uptake of the programs! That was an example of how we used Environics or demographic and sociographic data in order to help them understand a little bit more about the communities they were going to.
InsightaaS: So in that case, the ‘whole solution’ for the customer relies on a combination of the customer’s internal data, third party data, and your software, which allows the information to be organized in a way that's meaningful to the business objectives of the customer. Is that fair?
Kaszuba: Exactly. The key wasn’t what our software did on its own, it wasn’t what Union Gas had in their customer databases alone, and it wasn't what we could get from companies like Environics or from MPAC or other organizations. It's how it all came together that made the programs efficient. And in the end what allowed them to connect it all together was location: they can look at the physical location that they’re targeting and put these factors together to put together to drive the promotion and campaigns.
We refer that to this as the location based ecosystem. Once you’re sure that the core company information is accurate, and you can start to tie in these other pieces and have the same level of confidence, all of a sudden the models become more predictable and understandable, and they’re usable by business people. So we've developed multiple partnerships. Environics is one for demographics, sociographc and targeting perspective. They’re really valuable in marketing efforts. We also have a relationship with a company that has a database that's environmental issues from coast to coast. Our insurance and banking customers are very interested in that because if they look at a property and they can determine the proximity to a known environmental issue is within a certain range, they may want to have an underwriter go have a look at the property to make sure that the risk on that property isn't higher than what they assumed it to be in that model.
InsightaaS: How have you seen location evolve?
Kaszuba: As one example — we've done a lot of work with telcos around network planning. In the early days — in the 1990s — we would have been focused on answering questions like where is the best place to put your cell towers in relationship to where the population is, where's the traffic going to be, where will services need to be provided. Today with telcos, we're working on everything from network planning — trying to help them understand the optimal ways to lay down their network and make their services available to businesses — all the way through to sales and marketing: how do we help you target the businesses that are closest to your assets?
InsightaaS: In many ways that’s part of the dream of location based services, that ability to filter reality through, as you put it, the anchor of location.
Kaszuba: I think Deloitte did study a couple years ago saying that if you look at corporate data today, 85% of it can be anchored based on location. Most of the times it's an address — and one of the things that we've seen, and one of the things we really focus on addressing, is that the quality of that addressing data in most companies is not very good. And for lots of reasons, we've accepted it — it’s not great but it’s what’s in our CRM or, we know that our invoicing systems are out of date but the customer still pays the bill… But we know that with good address recognition technology, you can do a much better job. That's the non-sexy part of what we do, but it's critical. Without really strong, accurate information, it's hard to build good models.
InsightaaS: What do you find is a key driver for customers making the leap to location-based services? Are they justifying based on cost or environment or service delivery? What leads them to say this is a good investment?
Kaszuba: For government it has been around service delivery. A great example is the RCMP. They have a coast-to-coast emergency response program they call Safe Schools. It relies heavily on geospatial data to answer questions like, in an emergency, where do we put people? Where do we evacuate to? It's all very location oriented.
In the telco space the main driver has been network planning: how do we go into markets and do the most efficient job possible? We've done a lot of work with Wind Mobile, with Rogers,and with Bell, identifying up and coming communities. Where is the new building happening, and where do we put our dealers to service those communities or where do we put the cell tower to make sure we're servicing those communities.
On the banking and insurance side, it's a combination of things. One is the need to provide great service, responding quickly and accurately. If someone is sitting in front of me and asking for a mortgage, how fast can I get to yes with them and have a level of confidence that ‘yes’ is the right decision? It's also about efficiency. If I can use location information to avoid spending $500 or $1000 to have an inspector go to a property, I can save money for my organization. And what we’re seeing a lot lately is that banking and insurance is focused on managing accumulation as it relates to risk. One of their priorities — one that we're trying to help enable — is, how do I build systems so that I can look at a potential incoming opportunity, a new policy, and determine right off the bat what's my accumulated risk look like in that community today? Do I even want to be going after this business? And if I do, then I'll be the most competitive, I'll really drive hard to get it because it makes sense. If I don't then I'll adjust my policies based on that.
We did a lot of work with our insurance and banking customers around the Alberta floods as that was occurring. It was really interesting. With First Calgary, a credit union, the first thing they were focused on was, which clients are impacted? I need to know that right now because I want to reach out to them and ask how they are doing, and to let them know that we can help them while they're focused on their home — they don't need to worry about making a mortgage policy payment. And so they used location-based information to deliver great customer service, but then they also went on to use it to build analytical models to do all the things that a bank would have to do as a natural disaster was happening. That’s an example of how a real-time perspective provides a different level of value than what people have seen from location services before.
So it really has varied by vertical which is one of the things that has been exciting for us, the market for location isn’t limited to a specific vertical. It's very much a horizontal application. For years people have access to GIS data but it's been a backroom function and what we're trying to do is make it a front room, front office function that allows you to drive more efficiency in the business.
InsightaaS; It seems like there are cases where location provides a discrete solution, and others where it is integrated into other processes or applications. Can you provide an example of how you location becoming integrated into common business technology use?
Kaszuba: One — and we’ve talked a bit about this — is using location to drive better marketing and go-to-market campaigns. Could you build optimized territories utilizing location information so that your sales team is focused on the right customers, the right market opportunity? Location can have a huge impact on how you do that and how you build that.
Another example is managing operations and risk. If you can understand location and its proximity to things that are either positive influences or negative influences, you can build better risk models. You can build better models for servicing the clients you're going after. A simple example is flood and waterways. How close is this property to a waterway, and does it matter? A great example of that would be Scarborough with the Scarborough Bluffs. You might be close to water but the chances of it causing a problem would be pretty low. So all of a sudden elevation becomes part of the conversation but if this property is close to water and it has five other factors that may not be positive, you just might want to do more due diligence and drive a better process to make sure that you're managing it properly.