July 26

Contents

Editor's Introduction

This week, I propose that geospatial professionals take the lead in establishing public GIS centers and I report on how geoVue has been helping Papa Gino's decide where to locate new restaurants and whether to relocate any of its old ones. Plus, 21 press releases.

— Matteo Luccio

A Modest Proposal: Public GIS Centers

Just like war is too important a subject to leave only to the generals and law too important to leave just to lawyers and politicians — insert here your favorite examples of subjects that should not be left solely up to the experts — citizens should become more involved in planning land use and disaster response in their communities. As a former policy wonk, I understand that broader participation in the making of public policy tends to yield more practical, equitable, and stable solutions. So, merging my former and current passions, here's a challenge to geospatial professionals: they should take the lead in the creation of public GIS centers in their communities that would provide the hardware, software, data, training, and support to enable citizens to analyze local planning issues.

Such centers should be set up as independent non-profits and supported by funds from state and local government, local businesses, private donors, and low, sliding-scale user fees. Hardware, software, and data vendors should donate relevant products. Publishers of geospatial publications should donate subscriptions. Volunteers should be trained to assist walk-in users. Universities and community colleges should contribute instructors — in GIS, of course, but also geography, political science, economics, geology, ecology, agriculture, and other relevant subjects, depending on the most salient local challenges, as prioritized by the center's board of directors.

Here are a few examples of the types of questions that such centers might help citizens explore:

• How would a proposed new highway or subdivision affect wildlife? Housing prices? Traffic?
• Where is new public transportation most needed?
• Where are the poor concentrated and what does it suggest with regards to access to jobs, the location of child care and social services, and housing prices?
• What is the current mix of land use — residential, commercial, industrial, agricultural, recreational, etc. — in our city? Is it the most sustainable and equitable use?
• If, within twenty miles, a train full of chlorine derailed and tumbled down an embankment, a dam on a river breached, or terrorists crashed an airliner into a nuclear power plant, who would be affected? How would they be notified? Where should they go? How would they get there?
• If a tsunami hit near a coastal town, what areas would be affected, how fast, and what would the escape routes be?
• In case of a major earthquake, what major infrastructure facilities — roads, bridges, dams, power plants and distribution networks, etc. — would be damaged and what areas would, in turn, be affected?

I could easily come up with hundreds of examples, but you get the picture. I think the board should pick one hot local topic, to launch the center. Here in Eugene, Oregon, I would propose a six-month study of downtown revitalization. I would start with a presentation about the issues involved, followed by a two-week introductory GIS class (held in the evening, with child care provided), then weekly meetings of a study group. Members of the study group could drop by the center at their convenience — which means it should be open and staffed evenings and weekends — to work on the project. The goal would be to come up with three or four alternative proposals, illustrated with maps and data, to be presented to the City Council. Such a concrete goal, with a firm deadline, would keep the group motivated and on task.

Individual residents would be encouraged to come in to pursue their personal interests and projects — subject only to cost-recovery charges for printing and simple time rules to give everyone an equal chance to use the resources. The board of my homeowners' association is currently negotiating with a wireless carrier that wants to place a cell phone tower on our property. A board member or other volunteer could map the locations of other cell phone towers in the area, then contact the owners of the properties on which they are located to try to find out how much they are charging the cell phone companies in monthly rent. That sure would help us in our negotiations!

The centers could have as few as two computers but, of course, the more the better. How about twenty or fifty? Also, the greater the variety of software available the better — from the most basic viewers to complete software suites. At least introductory classes in GIS should be offered on site and those interested in intermediate and advanced classes should be referred to the nearest ones offered. Alternatively, if none is offered within a certain radius and there is enough demand, the center could bring in an instructor to teach a class. A library of GIS and public policy books would also be very helpful. Oh, and make the place as friendly and comfortable as possible! A furniture store could donate chairs and desks, an office supply store could donate paper for the printer, a local bakery could contribute a daily basket of goodies, and so on. Each donation should be appropriately and visibly acknowledged.

Monthly lectures and other presentations and demonstrations, announced in the local papers and on local radio programs, would draw people to the center. Of course, this would require a large meeting room. A good time to inaugurate a center would be on GIS Day. The next one is less than four months away, on November 14 — so you better call the first meeting of your GIS Center Founding Committee for next week!

geoVue Helps Papa Gino's Locate New Restaurants

One of the most common commercial applications of GIS is to help in choosing retail locations. Last week, geoVue — a provider of dynamic location optimization software for retailers, restaurants, and other real-estate based consumer channels — and Papa Gino's, Inc. — parent company to Papa Gino's Pizzeria and D'Angelo Grilled Sandwiches — announced that the latter is employing the former's software solutions to assist in its growth strategy for its two restaurant chains. The two companies, which have been working together for more than a year, are completing a plan that will help Papa Gino's prioritize markets, determine how many new restaurants to open in which trade areas, and how to rank them.

Papa Gino's has more than 170 restaurants throughout New England and in Orlando, Florida, while D'Angelo has more than 200 company-owned and franchised locations. Over the next five years, the two chains, which have been present in New England for more than 40 years, want to expand along the Eastern seaboard, in Florida as well as the Carolinas and New York. geoVue develops predictive store trade areas that help clients decide which markets to enter, expand, or exit; how they should optimize store networks within each market; and how to localize marketing and merchandising. Its modeling and optimization solutions help its clients plan store networks and prioritize capital investments based on predictive, multi-factor trade areas.

Using insight from geoVue, Papa Gino's evaluated all of the markets in the eastern half of the United States to determine which ones would work best for its brands. The company then determined locations within each market by assessing each of the retail areas within a market and then evaluating factors such as demographics, competition, and the expected demand for brands in the area.

I discussed this project with Anthony Padulo, Papa Gino's Senior Vice President of Franchise Development, and with Jim Stone, geoVue's founder and Chief Development Officer.

Interview with Anthony Padulo

Papa Gino's Senior Vice President of Franchise Development

1. What were you trying to achieve?

   We were looking to develop a strategy that created growth opportunities from a franchising perspective as well as continue our own corporate growth within New England. I had some familiarity with this kind of software because of my prior experience at Dunkin Donuts, where it helped us better understand trade areas. For me, the objective last year was to develop a program that would accomplish three things:

   1. tell me where we should be growing (which new markets outside of New England would be best suited for our products),
   2. help me understand what the potential is within those markets and where we should be located specifically, and
   3. help me do a sales forecast of that potential location, to help me decide.

   I also wanted to be able to do the same kind of analysis with our existing stores. Many of our restaurants are 30 or 40 years old and may not be in the right trade areas today.

2. What were the phases of this project?

   The first one was market ranking, which we completed last July. That's how we determined what markets to target. The second one was trade area analysis, which we completed last August and then tweaked through the fall. We are now completing the third phase, sales forecasting.

3. What are the key variables for you in picking locations?

   For me, the logical thing to do is to first decide what markets we want to be in; from there, what trade areas within the markets; and then the success parameters for the sites. So, the first thing that we did was surveys in our existing restaurants — to understand whether our customers go there from home, work, or somewhere else and where they go once they are done with us. Then, as we looked at the market ranking analysis, some of the key additional factors for us were the competitive environment, from both direct and indirect competitors, and the cost of entering a new market. We prioritized markets based on the cost of advertising in each one. Using GeoVue data, we also looked at such typical demographic factors as family size, population growth, and average income.

4. How did you select GeoVue?

   We looked at four or five different, well-established companies that offer the same type of information or services and narrowed it down to three: Claritas, MapInfo, and GeoVue. We worked closely with each of those companies over a couple of months to see which one was best suited for us. What we liked especially about GeoVue was that they would provide us the software that enabled us to do the analysis. Instead, Claritas and MapInfo were offering one-time analysis and then, if you needed more analysis, you had to go back to them. So the cost, over time, would be significantly higher. More importantly, geoVue gave us more flexibility in looking at different scenarios. We also believed that GeoVue offered us the best service and follow up. Last but not least, they were also local, which we felt was especially important for us, being new at this.

5. What in-house GIS expertise did you have?

   We had no GIS expertise in house. We hired a market planner who had done this type of work at TJMaxx and had four to six years of experience working with other programs. He had the technical expertise, but was not acquainted with geoVue. Over the past year, he has worked with geoVue, which has provided the training, and he is now doing an exceptional job of managing and working with the program.

6. What do you wish to achieve in the next few years?

   I wish that in three years we will have demonstrated that the program works and we will have been able to reach our new development growth as well as our current location analysis. If we meet those criteria we will be very, very happy.

Interview with Jim Stone

geoVue's founder and Chief Development Officer

1. What do you offer your customers?

   One of our objectives is to enable them to perform analysis themselves, in-house, at whatever level they want to operate. Some are very sophisticated, and have large in-house staffs. They are looking for software, data, and training, but other than that plan to be very self-sufficient. Other customers are either relatively new to using these methods and tools or don't have large staffs. One of the components of the solution that Papa Gino's is using is a market planning and optimization tool, called iPlan. It requires a fair amount of experience to use and Papa Gino's decided to hire an analyst who was able to do so. We have other customers who may not open enough stores and may not want to make the investment in a person who can do that. We will then provide the market planning as a service to them and tell them which markets they should focus on and how many stores they should put in each. Then they can use that as a starting point for their planning and site selection process.

2. How did you develop your expertise in this area?

   Over a year ago, the head of research for Albertsons and several people from her staff joined geoVue, so that we have a level of expertise in-house that is pretty much unmatched by the other companies that do what we do. These are people with twenty years of experience working at one of the top and largest chain operators in the United States, who understand what it is like from the client side of the fence. That helps us deliver services, as well as analyze the requirements of our existing and potential customers to make sure that we are focusing on the things that are really relevant — as opposed to creating lots of cool software that nobody can use.

3. Do you provide the data, get it from your customers, or buy it from third parties?

   Every solution that we develop is a combination of all three of those options. We have access to all the third-party databases, whether it is demographics or business locations, and all the cartography that is necessary to set up systems. Additionally, we can use data that our clients have in-house — including their store networks, sometimes their own competitor databases, and customer data. Sometimes, as was the case with Papa Gino's, we need to acquire additional proprietary data, in order to build reliable models. That might require doing customer intercept surveys or process point-of-service data that is somewhere in the bowels of the client's IT department.

4. What kinds of analyses do you perform?

   We use two broad classes of analysis. The first one is aggregating data to trade areas or markets and then looking at the composition of those trade areas, using regression models and other techniques, to observe trends in the data and understand what drives sales. The key is to define these areas in a way that is appropriate for understanding the way customers actually make choices. For a restaurant chain, such as Papa Gino's, the customers are coming from a combination of their homes, work, or shopping. We need to understand these trip patterns and the demand in the area. For example, is a location near many offices and some residential areas with large concentrations of families? The second one is a gravity model (better known in GIS as a spatial interaction model), which we use extensively. It allocates the demand in a market to the different potential supply points — which is to say that we are modeling how consumers are choosing among the alternatives that they have. Direct competitors are restaurants that offer similar products and go after the same types of costumers; indirect competitors sell another meal that might compete with Papa Gino's menu. It is basically computing travel distances in order to predict the probability of patronage. We simulate all of that through the road network and then sum up the expected models at the various supply points to try to predict what the volumes are going to be. We are then able to run scenarios that allow the user to see the effect on his total revenue of putting stores in certain locations.

5. What did you learn working with Papa Gino's that will help you with other clients?

   Papa Gino's has made a commitment to use quantitative analysis to try to improve their decision-making. This is something that really starts at the top of the company — with the CEO, the CFO, and Tony Padulo. The senior executives have really embraced this approach to try to make better decisions and to extend their presence beyond their traditional New England market. That in itself is not as common as you might think. The second thing that happened, as a result of us being fairly local, is that we had the opportunity to spend a lot of time together sorting out exactly what components of analysis are relevant and more helpful than others and working with senior executives on that. As a result, geoVue developed a better sense of what executives consider to be important as opposed to what we geeks on the vendor side think is cool and powerful. They expanded their horizon in terms of the power of some of these predictive analytics tools and we learned better ways to make it understandable and relevant to the decision-makers.

6. In what direction are you moving with this technology?

   We are moving to a more Internet-based delivery system. We are getting ready to release some new Internet-based solutions later this year that are going to make it much easier for our customers to update and make changes to data as well as to some of the metrics of the models. This will allow them to keep their systems up to date more easily and it will allow us to extend access to these maps, reports, and analyses beyond the real estate department and the techie people to other folks across the enterprise.

7. Are the systems you delivered to Papa Gino's server-based or desktop-based?

   They are all desktop-based. Many of our more than one hundred customers are using these solutions the same way. However, we've also had Web-based applications for ten years or so. In the last three or four years, we've increased our focus on improving our analytical applications to the Web-based environment.

8. Do you envision moving all of your applications to the Web?

   No. There will always be a need for desktop applications, because power users want to be able to do things that are just not very practical to do in a Web-based environment. They are building things that they can then publish to the Web for others to use. We refer to doers and viewers: the doers are the power users who create content output for the viewers to use. The doers need the desktop environment in addition to the Web, whereas the viewers just use a Web browser to look at the different maps and reports and get answers to questions that they need in that environment.

» Back to our GIS Monitor JUL 2007 Issue