STI: PopStats 2014 Research Conference Report

From a Client Case Study on Wednesday Morning to Robert's Annual Deep Dive into Data Methodology on Friday, the 9th Annual PopStats Conference Delivered the Ideal Balance of In-Depth Data Education and Peer Networking - All With a Stellar View of the Rapidly Changing Skyline of America's Fastest Growing City.

This year's PopStats Research Conference was held from Wednesday March 19 to Friday March 21, in Austin, Texas, on the 17th floor of the Hyatt Regency Hotel on the shores of Town Lake - with the large and expanding city skyline of the nation's fastest growing city in full view. The stunning location gave attendees convenient access to the much-lauded Austin downtown food, music, shopping, and entertainment scene.

As always, the 9th Annual STI: PopStatsT Research Conference and User Forum was all about the users of the Synergos Technogies' data. In fact, STI is not just a client to the users, it's also a customer of many of its clients - as Lesley Newman pointed out in three different scenarios in her introduction: "I started thinking about this conference last year while preparing for a sport viewing event at my home - from the extra television we rented at Rent-A-Center to the snacks we bought at Whole Foods. When I had a medical emergency, I went to Walgreens for my medications. To prepare for this conference, we relied on many of our clients' products and services - from printing and supplies at Staples to sustenance fortification on Robert's `Chipotle Wednesdays'." And that is just a few of the many PopStats clients from whome the STI team regularly consumes products and services.

Feedback from Conference Attendees!

Every year, the feedback from conference attendees helps us understand how we did this year - and how we can do better in future user conferences. We're proud to say that we receive a lot of great feedback. We're happy to hear that we are meeting attendees' information needs and expectations. Almost 94% of the attendees said they will attend the conference again next year, and 100% said they will recommend the conference to others. Here is more attendee feedback from the conference surveys:

How would you rate this conference compared to other conferences of this type that you have attended?

• Much more content and real examples.
• I think this conference is better put together. It's organized, professional, and friendlier than many of the other conference I've been to.
• First of its kind.
• Great content
• Way better than average.
• We prefer this conference over others because it tends to be more educational then others.
• Good mix of different companies and sectors.

What was the single best idea you gained this week that will help you in your job?

• Chain our market optimization model to run successively for 1 location, 2 locations, 3 locations, etc., as a way to examine for improvement.
• Website searches.
• To build analog models.
• Ideas for expanding GIS use.

What new topics did you hear about that you plan to investigate further?

• Online systems and Tetrad's technology offerings.
• Segmentation.
• Looking at trade areas in ways to reduce the number of non-customers in that area.
• 10-year projections.
• Analog models and market planning through seed sights.
• Some of the insights from the researching on the internet session.
• Some of the post-motion measurement ideas from David Chipman's session.
• Reverse appending street addresses using a hybrid model for land and cell phones.
• WalMap (a Walgreens concept).
• Being able to use the data to help prioritize opportunities.

This Conference Report provides an overview of the 2014 Research Conference, including synopses of all the sessions. This Report provides a review for this year's attendees, as well as an overview for those who could not attend the event.

TABLE OF CONTENTS

Wed. March 19, 2014

SESSION 1 - Answering All the Questions, Brian Quinn, Tractor Supply
SESSION 2 - New Concepts - Targeted Market Approach, Ross Wootton, Chipotle Mexican Grill
SESSION 3 - Lipstick on a Pig: Modeling Assumptions and How Models Can Be Manipulated, Paige Stover, Forum Analytics
SESSION 4 - Researching on the Internet, Kyle Day, STI
SESSION 5 - Looking to the Past to Predict the Future, David Chipman, Men's Warehouse

Thr. March 20, 2014

SESSION 1 - Top Growth Markets for 2014, Robert Welch, STI
SESSION 2 - Preferred Market Analysis (PMA) Econometrics, Maxwell Fisher, STI
SESSION 3 - A Metaphysical Conversation on the Nature of Trade Areas, Joe Rando, Trade Area Systems
SESSION 4 - "WalMap" - Building a Web-Based GIS for Walgreens, Ben Farster, Walgreens
SESSION 5 - An Insider's Guide to Market Optimization, Devon Wolfe, Tango Management
SESSION 6 - Bridging the Gap Between Market Strategy and Real Estate, Lorie Williams, Family Dollar

Fri. March 21, 2014

SESSION 1 - STI PopStats Methodology, Robert Welch
SESSION 2 - STI LandScape Methodology, Robert Welch

SESSION 1 - Answering All the Questions

Presenter - Brian Quinn, Tractor Supply

Where do the nation's recreational farmers and ranchers with one to 25 acres live? This is the central question driving Tractor Supply's GIS department, Brian Quinn told attendees at the first session of the 2014 PopStats user conference. Consequently, unlike most companies, Tractor Supply hunts for trade areas in America's rural areas. As a result, the company doesn't need a large number of residents as much as it needs rural residents with certain income levels.

However, it's not the only question Tractor Supply continually asks about its 1,250 and growing locations in 47 states. Additional questions are:

• How many locations can we open?
• Where are we opening next?
• Does this feel like a good site?
• How are competitors impacting us?
• How accurate is the forecast?

"We hit a wall a few years ago on finding locations in rural areas, which is fundamentally more difficult than urban and suburban areas," explained Brian Quinn. "We needed a new strategy that answered all of our critical location questions." Adding to the challenge is the company's goal of reaching 2,100 stores.

Tractor supply has gained better answers to all of its location questions since implementing more in-depth predictive analysis in the past five years using a combination of internal and external data and a new modeling process that includes these five steps:

1. Identify site
2. Make an initial prediction
3. Make an impact prediction
4. Compare to an analog store
5. Make a forecast

Attendee Feedback:

• Awesome!

SESSION 2 - New Concepts - Targeted Market Approach

Presenter - Ross Wootton, Chipotle Mexican Grill

Unlike the majority of QSR, Chipotle rarely adds new items to its menu. With only a couple of exceptions, Chipotle offers the same small selection of food choices today that it offered when it opened its first store in 1993. However, unlike most other QSR's Chipotle is considering expanding into new store concepts with distinct food items. These include Pizzeria Locale with Neapolitan-style Pizza, and ShopHouse Southeast Asian Kitchen. What's more, the company is considering new location options, including food courts, airports, and military sites.

To test these innovative concepts, the Chipotle research staff was asked to perform a scan of specific markets across the U.S. with one goal: To find how many potential trade areas exist for each new food concepts. To accomplish this feat, Ross Wootton used a hexagonal gridding concept presented by Andy Moncla during his presentation at last year's PopStats conference.

Ross ran two scans - one he calls 20/20 and one 10/10. For example, the 20/20 scan included:

• Home populations over 20,000
• Daytime populations over 20,000
• Two-mile trade areas

The 20/20 scan netted 654 trade areas that matched the criteria: 244 in the west, 185 central, and 225 in the northeast. We'll see where new Chipotle, Pizzeria Locale, and ShopHouses open in the future.

SESSION 3 - Lipstick on a Pig: Modeling Assumptions and How Models Can Be Manipulated

Presenter - Paige Stover, Forum Analytics

Paige Stover presented conference attendees with an argument on the downside of striving for 100% accuracy when conducting location research. Theoretically, a company's executives may say they want 100% accuracy for market research. But there is much to be lost in placing such an unrealistic demand on a company's site selection team, says Paige Stover. Some companies seem to believe that you just "look at the best stores, find their commonalities, and do that over and over. As a result, we often find ourselves in the position of educating our clients on the art of how markets work and the science of modeling."

But if you strive for accuracy in modeling location, "you'll miss interesting information," says Paige. She cited the following examples:

• Low profit stores are too close to each other
• Opportunities missed to remodel or reorganize low to medium performing stores
• Opportunity to find stores that are over-performing in a market

"Our goal is not to achieve one hundred percent accuracy with our models, but to help our clients make the best decision possible with the data," noted Paige.

SESSION 4 - Researching on the Internet

Presenter - Kyle Day, STI

Conference attendees enjoyed a crash course from the youngest full-time member of the STI team - Kyle Day. He presented a long list of both traditional and non-traditional sites where researchers can access interesting and valuable research on industries, markets, companies, and products. He not only provided the research sites, but also demonstrated how to access relevant data, including a few web research tricks and tips. Here are just a few of the dozens of internet sites that Kyle shared with attendees.

• NAICS Codes - multiple levels of information access
• Scribd - Ebooks and documents access
• SlideShare - Presentations access
• FullTextReports - Detailed academic reports
• JSTOR - General information reports
• FRED - Econometrics data
• BizNar - Business journals access
• WebWire - News by industry categories
• GovScan - Local news and government activity
• ZanRan - Semi-structured industry data
• MarketVisual - Finding executives and associations
• ZoomInfo - Finding people and associations
• SocialMention - Finds what's on people's minds
• WayBack Machine - Archives of old websites
• Alexa - Website traffic data

• Well-spoken young man.
• Good for you!
• There was a lot of relevant material.

SESSION 5 - Looking to the Past to Predict the Future

Presenter - David Chipman, Men's Warehouse

David Chipman concluded the first day of the PopStats conference by sharing some of Men's Warehouse's biggest market research challenges - from competitor impacts to trade area capture rates to analog selection - and discussed how he overcame them by looking back at the company's growth history.

First, he overviewed the company's major forecasting challenges including:

• Is the trade area accurate - or is it too big or too small?
• Are the capture rates too high or too low?
• Are the competitor impacts more or less?
• Are the maturity curves accurate for all stores - or do some mature faster or slower?

Then David gave examples of what the company discovered in post-forecast analysis in several markets including the following:

• Newton, NH - Men's Warehouse grabbed a bigger share of the market than predicted. Post-forecast analysis showed that the area lacked in competition.
• Hawaii - Got a lower market share that predicted at above-average. Post-forecast analysis showed that the company missed including a local analog store in its research.
• Las Vegas - Under-predicted sales. The data Men's Warehouse was using at the time gave a population count that was 100,000 lower than PopStats accurately provided in post-forecast analysis.

SESSION 1 - Top Growth Markets for 2014

Presenter - Robert Welch, STI

Robert lead off the second day of the conference with his annual presentation on the top growth markets across the U.S. This is always a favorite session - particularly since the research also shows top market growth since he started the research eight years ago. Robert overviewed the top growth markets and pointed out major changes from last year - most notably, the move of Midland-Odessa, Texas, to the number one position, bumping Austin to number 2.

Robert uses the same methodology each year to create the report - which makes it a valuable tool for year-to-year comparisons. The criteria used to create the list includes areas that: (1) have an initial market size of at least 60,000; (2) have a minimum per annum growth rate of 2%; (3) have a minimum growth rate of 8,500 people; and (4) experienced significant growth for at least two years. Robert uses PopStats data for the analysis.

This year's report showed 27 top growth markets that met the research criteria. The top 10 are:

1. Midland-Odessa, TX
2. Austin, TX
3. Charleston, SC
4. New Orleans, LA
5. Raleigh, NC
6. Houston, TX
7. S. Jacksonville, FL
8. E. El Paso, TX
9. Dallas-Fort-Worth, TX
10. W. Orlando, FL

Attendee Feedback:

• I like this topic every year - and I like that you remind us how accurate your population count is compared to the U.S. Census.
• Don't know how you would improve it, but definitely keep doing it!
• Always a favorite session.

SESSION 2 - Preferred Market Analysis (PMA) Econometrics

Presenter - Maxwell Fisher, STI

Maxwell Fisher gave a presentation on the new Preferred Market Analysis (PMA) research STI is going to produce with STI: Colossus econometric data. He presented an overview of how the Colossus data can be applied to a PMA project as a way to leverage the massive database and make the data more accessible to market researchers.

"Colossus has 135 unique data sets and 28 million variables, with both traditional data (like GDP and unemployment) and non-traditional economic data (like weather and crime)," notes Max. The benefits of creating PMA econometric reports include:

• Enhances users understanding of top growth markets.
• Explains the economic health of top growth markets.
• Shows the economic resiliency of top growth markets, which is calculated by dividing the local unemployment rate by the national unemployment rate.

STI's upcoming PMA reports will contain essential data from several key Colossus econometric datasets, for example (partial list):

• Building Permits
• Local Area Unemployment
• Construction Growth
• Education and Health Services Growth
• Leisure and Hospitality Total Employment
• Manufacturing Growth
• Professional and Business Services Total Employment
• Financial Activities Job Growth
• Trade, Transportation, and Utilities Growth

SESSION 3 - A Metaphysical Conversation on the Nature of Trade Areas

Presenter - Joe Rando, Trade Area Systems

Joe Rando turned his presentation into a lively discussion with the conference attendees on how to create trade areas, strategies, density, and dealing with competition. The format allowed everyone to learn from other researchers' experience in these areas.

One of the topics was the issue of trade area noise. "At some level, you can only focus on the people inside the trade area who will actually shop at your locations. The rest is noise," said Joe. What's more, sometimes the more market share a company tries to grab in an area, the more noise they'll get. "The goal is to get the most target customers with the least amount of noise. It's a balancing act with a Goldilocks Effect - not to much or too little, but just right."

Attendee Feedback:

• Great!

SESSION 4 - "WalMap" - Building a Web-Based GIS for Walgreens

Presenter - Ben Farster, Walgreens

With 8,500 stores in 49 states, some operating under different names, Walgreens needs more than just a real estate research team to find new location - it also needs research tools that are accessible across the enterprise. Ben Farster is Manger of the company's Enterprise Location Intelligence team. The team spent five years creating a web-based GIS application called WalMap, which allows key departments to gather information on existing store locations, target new opportunities, monitor competition, and more. Ben overviewed the creation and use of WalMap in his presentation.

Objectives for WalMap are to support Walgreens' three key strategic growth drivers:

• To deliver a "wellness experience"
• To encourage pharmacist to communicate more directly with customers
• To establish an efficient global platform

Walgreens first launched a version of its web-based GIS system in 2006. The company updated the concept over the past five years and released an updated version in 2013. In March 2014, the company launched WalMap Pro and will have 500 to 1,000 users. The next step is to launch a mobile version of the app. Creating the application involved the following key tasks:

• Conducted a user needs assessment - which included information like store locations drawing trade area maps, and attribute editing
• Conducted a general needs assessment - such as determining where a location is in a market and where the company wants to go
• Created a report that solidified the technology development roadmap
• Obtained a realistic budget - part of which included the management stipulating that the app integrate with other applications used across the enterprise

Attendee Feedback:

• Great - particularly the challenges involved with rolling out an enterprise solution.
• These guys are the best.
• Always impressed.

SESSION 5 - An Insider's Guide to Market Optimization

Presenter - Devon Wolfe, Tango Management

Devon Wolfe made a strong case for research optimization in his presentation on Thursday afternoon, including why do it, how to do it, and how to run it. He also presented trade area research scenarios and how they would have turned out better with optimization. During his 15-year career in market research, he's observed that the two biggest problems with optimization are:

1. There is no standard for defining optimization
2. Most so-called optimization processes do not get to true optimization

Devon explained that his definition of optimization is: An ordered search of possible site opportunities using defined criteria.

"You ought to be optimizing your markets," he said. Then he provided a list of reasons why:

• Gives you a qualified starting point for your real estate research efforts
• Frames your decision-making in a broader content
• Allows you to rank and prioritize among multiple markets
• Provides a defensive target for a chain
• Exposes diamonds-in-the-rough that might not otherwise be discovered

SESSION 6 - Bridging the Gap Between Market Strategy and Real Estate

Presenter - Lorie Williams, Family Dollar

At one point, Family Dollar's growth was between a rock and a hard place. The company was caught between an old-school field research system and a need for an optimized site location system. "Our field team was still using on a gut-feel and shotgun approach to site location," says Lorie Williams. Compounding the problem was the company's site location structure, which is built upon having about 60 real estate managers in charge of their territories.

The solution was to create and deliver multiple market research tools to empower the company's real estate managers in the field. "Our goal was to give the managers the tools they needed to make the same decisions we would make about site location," explained Lorie in her presentation. "An additional goal was to determine what part of market research we could give the managers that would make us more efficient in the market research department."

To close the GIS knowledge gap between the market research department and the real estate managers in the field, the market research team created and deployed a web-based GIS application, which launched in 2012. The system achieved several goals:

• Research is based on logic and rules versus gut instinct and best guesses
• Everything is standardized
• Market research packages streamline the processes
• Guidance for scoring systems based on different scenarios are provided

Attendee Feedback:

• I really enjoyed hearing how research and real estate are on the same team.
• No wonder they are such a strong company.

SESSION 2 - STI: PopStats Methodology

Presenter - Robert Welch, STI

Robert rounded about the two-and-a-half day Conference with explanations of the PopStats' methodology, the five-year forecasts, the 10-year projections, and PopStats Expected Value Index. Regarding the methodology, Robert said that the PopStats estimates are calculated on multiple computers working together in a pseudo-parallel processing manner. "We have created a self-correcting artificial intelligence modeling system that learns from itself." Here is a brief description of the methodology. Four models comprise the PopStats' model:

• Zip+4 Model. This model is based on over 28 million ZIP+4s, representing over 116 million households. This model is the primary determinate in understanding population growth and decline. Vital to the process is that STI maintains its own street files that feed into PopStats, because "we do not want spurious third-party data entering into the calculation."

• Postal Delivery Model. This model is based on postal delivery statistics provided by the postmasters in each market across the country. This model's primary purpose is to understand trends in existing populated areas.

• Spread Model. This model, which is based on macro-level postal counts, performs double-duty: It both calculates populations in rural areas where ZIP+4s are limited, and serves as a checks-and-balance for the previous two models.

• Census Model. This is the grand master of all the models, which STI also calls the "black box." It pulls together the other three models using an extreme set of heuristics (if-then questions). In short, it is the final decision-maker of the estimate.

The PopStats methodology includes automated processes for overcoming any and all anomalies present in the data, including ZIP+4 inaccuracies, data smoothing issues, conversions (lofts), and overrides. If a client has a question about PopStats' estimates, Robert will research the issue. The client should first gather as much detail as possible, include any third-party evidence that supports his or her position, and be patient during the evaluation process. Robert will verify the issue, check the underlying data, check soft sources, and report back to the client in a timely manner.

Attendee Feedback:

• Best lecture of the conference.
• Extremely helpful in explaining to clients and superiors how and why PopStats is superlative.
• The information presented is fantastic; it enables me to confidently explain why the data is good and the nuances of variable types within the company.

SESSION 1 - STI: LandScape Methodology

Presenter - Robert Welch, STI

Robert added an in-depth discussion of LandScape's methodology this year, beginning with an overview of the four reasons that neighborhood segments consistently form across the country 80 percent of the time:

1. Tradition. A neighborhood has been one way for a while, such as nouveau riche or retired citizens, and it stays that way.
2. Perceptions. Residents' attitude is "if I lived there than that's where I perceive myself living," such as a hipster or working family neighborhood.
3. Development. Some neighborhoods are specifically designed to fit a personality type.
4. Physical Environment. People who share a similar lifestyle are attracted to similar geography, such as mountains, farmland, or water areas.

Neighborhood segmentation works because the segments change slowly. But one weakness is that while the demographics may indicate a specific consumer type, such as "married with children," the attitudes of the residents may be different that what is expected. For example, some parents may raise their children according to an attitude of ensuring they are always dressed in the latest fashions. Meanwhile, other parents' attitudes could be that "our children will just outgrow their clothing, so we're just going to purchase discount items." LandScape's ten social and lifestyle indicators add a new dimension, and a way to see these attitudinal differences, to the traditional breakdown by demographics.

Announcements