2009 PopStats Research Conference Summary

Attendees Enjoy In-Depth and Innovative Market Research Insights

Expected Value Index. Sparklines. LandScape's lifestyle indices. These are just a few of the many essential take-aways cited by attendees during the 4th Annual STI: PopStats Research Conference and User Forum held in Austin at the Stephen B. Austin hotel on March 25th - 27th.

Attendees' said they especially appreciated the knowledge-sharing and innovative insights offered at this year's Conference including:

• Hearing best practice case studies from retailers such as Family Dollar and Ace Hardware
• Discovering unique market research techniques
• Learning about new Synergos Technologies products under development
• Seeing a discussion of the STI: PopStats methodology from the product's creator
• Gaining the opportunity to network with peers

As a follow up to the event, we've prepared a User Conference Brief with synopses of all the sessions. This is both a review of the highlights for those who attended and an overview for those who could not attend. For additional information on the 2009 User Conference sessions, download the available session presentations.

TABLE OF CONTENTS

Wed. March 25, 2009

SESSION 1 - What's New and Major Initiatives Taking Place, Robert Welch, STI
SESSION 2 - Segmentation: It's All About Potential!, Larry Maves, Mapping Analytics
SESSION 3 - Quantifying Site Selection, Brian Strickland, Family Dollar
SESSION 4 - Capturing the Hispanic Customer, Dave Huntoon, Intalytics
SESSION 5 - The Top 50 U.S. Markets, Robert Welch
SESSION 6 - Finding the Silver Lining, Robert Welch

Thr. March 26, 2009

SESSION 1 - How to Squeeze 800K Numbers on a Piece of Paper, Robert Welch
SESSION 2 - Love at First Site, Paul Sills, Ace Hardware and Forum Analytics
SESSION 3 - Bringing it All Together, Wayne Geary, Jones Lang LaSalle
SESSION 4 - The Secret is in the Gaps, John LeTourneux, The Kroger Company
SESSION 5 - Extending the Market Research Dept's Insights, Dave Hanssen, CVS
SESSION 6 - You Buy What You Think, Cindy Reid, Mapping Analytics
SESSION 7 - Request for Enhancements, Robert Welch

Fri. March 27, 2009

SESSION 1 - STI PopStats Methodology, Robert Welch

ATTENDEE COMMENTS

• This is a great conference. I gained as much from the conversations as from the presentations. You give good value.
• Great job. I am very impressed with the speakers and the conference's overall execution.
• This conference continues to rate highly versus other similar conferences.
• Much more informative, less formal, and more intimate. Fabulous!
• Better, more relevant information regarding how the companies are using data in practical applications.
• I learned a great deal more here than at the ICSC research conference.
• Superior in every way.
• It was the best conference I attended this year.
• It was a great opportunity to speak with other users of PopStats data, and the user presentations were helpful as well.
• People that have attended in the past have said that this is the best conference.

SESSION 1 - What's New and Major Initiatives Taking Place

Presenter - Robert Welch, Synergos Technologies

Robert kicked off the 2009 Conference with an overview of STI developments and news since the last User Conference in 2008. Highlights of his presentation included the following points:

• STI Growth. Despite the economy, STI grew its customer base by 23% last year, including new clients from all trade industries. An independent ICSC study found that STI is one of two demographic data providers preferred by retailers. Also, three out of four retail panelists at an ICSC best practices session cited STI as their demographic data provider.
• New Data. STI: PopStats has expanded in the past year to include five-year forecasts on income, single-year breakouts on ages up to 21 years, mortgage risk data, five-year race forecasts, race diversity, and labor force and unemployment refinements.
• STI: BlockPoint. This product launched in the summer of 2008, providing researchers with block-level population estimates.
• New Products. Robert announced the 2009 launch of two new products:
• STI: Trade Zones. This is a workplace segmentation system that will include block-level segmentation of business and job categories including: retail, wholesale, service, medical, finance, manufacturing, public, and residential.
• STI: Market Outlook. This is an add-on to STI: Spending Patterns, which will provide insight into retail sales gaps.
• Client Award. The Louisiana Department of Social Services won an industry award for a new technology developed using PopStats demographic data.
• New STI Website. STI is launching a new website by May 2009 that will combine www.synergos-tech.com and www.popstats.com into one single online demographic data information destination. STI will continue to add valuable information and services online, such as last year's addition of STI: Data Depot, a free repository of a robust selection of GIS data, and STI: ISIS, a easy-to-use reporting software created by Robert.

Attendee Feedback:

• Good way to start the conference.
• Great ideas on where and how to move forward with data advances.
• Very helpful.

SESSION 2 - Segmentation: It's All About Potential!

Presenter - Larry Maves, Mapping Analytics

Larry discussed how complex market analysis can become in today's business world, yet how much easier it can be to conduct research by using powerful analytic tools. Thanks to more robust software and data, today researchers can conduct research they could not do before. Larry overviewed an example of the research possibilities for competitive market analysis using a consumer segmentation system. He mentioned that PopStats is much better than other data products at "finding the truth of what's on the ground. Plus, for some market studies, such as our Retail Signatures program, quarterly population data is essential."

Attendee Feedback:

• I think the subject matter is important.
• This is an interesting concept.

SESSION 3 - Quantifying the Importance of Site Selection

Presenter - Brian Strickland, Family Dollar

Brian discussed how revamping Family Dollar's market research techniques made it a much more viable tool in the company's real estate decision making. In fact, the retailer achieved improvements in the 3-5% range. The impetus for this move, says Brian, is that the company faced challenges with researching 800 to 1,000 sites a year, including expansions, relocations, remodels, and 200 new locations.

Brian mentioned that researchers dramatically improved research results in estimating market potential when they started using PopStats. Prior data underestimated household counts: as a result, the research underperformed. After updating all market models with PopStats, Family Dollar found that site quality was the leading factor in a store's success. Site quality includes population density, road networks, patterns of development, co-tenants, visibility, and access.

The new site analysis tool revealed how much better or worse each site would perform based on its site characteristics. In addition to new locations, Family Dollar scored all of its existing stores to see if they were worth keeping. Brian stated that "what we expected and what we found out was very different. The benefit was being able to confidently advise the company on site potential and what factors are necessary for a successful location."

Attendee Feedback:

• Brian did a great job.
• Very insightful.
• Very good presentation and excellent speaker.
• Very interesting approach.

SESSION 4 - Capturing the Hispanic Market

Presenter - Dave Huntoon, Intalytics

Understanding performance potential in individual stores is an ongoing challenge for many QSRs. Dave explained how one regional fast food chain better identified its store performance through a combination of demographic research and consumer surveys. Research for this project started with two primary questions: what drives individual store performance? And how can customer information translate into actionable insight?

"We found that dependable population estimates and neighborhood segmentation insight are critical components of store performance," said Dave. "Knowing how many bodies are in the trade are has a major impact on results, because retain and service demand is directly impacted by the number of potential customers. In our case, it involves 40% of the equation."

To better understand its largely Hispanic consumers, the fast food chain incorporated data from STI: LandScape's Espaniola category. Among the insights it gained are that higher versus lower income segments performed better and showed a little better results with older versus younger residents.

Attendee Feedback:

• The discussion was valuable.
• Good questions were asked.

SESSION 5 - The Top 50 U.S. Markets

Presenter - Robert Welch, STI

Using the same methodology as previous years "to provide accurate year-to-year comparisons," Robert revealed this year's top high-growth markets to Conference attendees. This valued document is exclusively available only to Conference attendees. The report revealed several notable changes from last year including:

• There are only 41 growth markets that matched the three criteria (see below).
• Las Vegas dropped to 39th place from number one in 2007 and number 17 in 2008. However, the 10-year projections find it returning to first place.
• New Orleans retained its number one position from last year. Much of the new population is Hispanic.
• Austin is the second fastest growing area in the U.S. and one-third of the growth is geographic.

Robert explained the methodology used to create the high-growth market list including how he segmented the markets and the fact that each market had to meet the following three criteria:

• Current population - must be over 60,000.
• Per year growth - must be greater than 2%.
• Growth projection - must be greater than 8,500 per year.

Attendee Feedback:

• Robert does a great job of presenting data and ideas.
• Very informative.
• Always an interesting topic and I like the year over year comparison.
• I always enjoy the top 50!

SESSION 6 - Finding the Silver Lining

Presenter - Robert Welch, STI

Today many companies want to know: In which markets are we vulnerable to economic calamity? Robert demonstrated a data value metric found with the PopStats data set to discover the economic health of a trade area - in particular areas poised for downswings or upupswings. This metric is called the Expected Value Index (EVI).

Robert demonstrated EVI using thematically shaded maps. In his examples, he showed how EVI can reveal not just growth, but growth type, including exponential, linear, and declining. He illustrated how researchers can determine at-a-glance exactly which areas are growing or declining as expected, faster than expected, or slower than expected. He noted that one of the benefits of EVI is that companies can "be on the bleeding edge and the first to enter into a fairly established market."

Robert also explained how to create these econometric analytic tools using PopStats data:

• Growth Stability Index. Used to find erratic, linear, and curved linear growth patterns.
• Housing Demand Score. Uses PopStats vacancy rate data.
• Mortgage Risk Index. Uses a ratio of gross income to loan value.
• Housing Value Change.

Attendee Feedback:

• Love to hear Robert talking about potential ideas.
• I always appreciate seeing new ways to use the data in an easy-to-assess visual.
• The EVI index is awesome! This is the type of information I'm looking for at your conference. Also very informative.
• Very interesting!
• Very helpful. My favorite topic of the Conference, because of its relevancy to the current market environment.
• This is a great value add. I really like to learn information I'm not familiar with.

SESSION 1 - How to Squeeze 800,000 Numbers on a Single Piece of Paper

Presenter - Robert Welch, STI

Robert introduced a high-density graphic method called "Sparklines," and discussed its potential uses by researchers to communicate a large quantity of data in a small space, especially to executives. Sparklines are defined as a type of information graphics characterized by its small size and data density. Sparklines present trends and variations associated with some measurement, such as average type or stock market activity, in a simple and concise way. More information on Sparklines can be accessed on several websites including:

• www.bonavistasystems.com
• www.bissantz.com/en/
• www.omnipotent.net/jquery.sparkline/

Attendee Feedback:

• I enjoyed this "science and art" session - and don't mind future sessions being more science/numbers oriented.
• Great idea on how better to show sales data and/or historical comp data for faster decision making that is requested from Sr. executives.
• I like this type of theory much better than statistical/math based in the past. This is really valuable because of links to websites makes it something I can use with little research.
• Great and interesting session.

SESSION 2 - Love at First Site

Presenter - Paul Sill, Ace Hardware, and Forum Analytics

Every month, Ace Hardware receives about 6,000 site opportunities from brokers, developers, and landlords. Out of that number, Ace opened 88 stores in 2008 and plans to open 105 stores in 2009. To reign in this unmanageable number, the research department and Forum Analytics created a way to quickly analyze this input to proactively filter the list. The primary goal was to find the best opportunities among all the options presented.

The new site analysis tool sets different risk levels for all variables that were identified as critical to success. "Our goal is to only pursue pre-qualified sites that are in optimal, high-interest areas," said Paul. "Our ultimate goal is to maximize the ROI of every dollar invested in real estate research."

Results of the analytic process show that units selected through this process do 5.2% better than others. "This tangibly validates the impact of analytics to the site selection process," added Paul. "This process also showed the company how our department can build models for strategic decision making as well as site selection."

Attendee Feedback:

• Great presentation. Lots of detailed content going on.
• The concept was great.
• Good presentation concept. Easy to understand.

SESSION 3 - Bringing it All Together

Presenter - Wayne Geary, Jones Lang LaSalle

To select ideal locations, Jones Lang LaSalle, a commercial real estate firm, found that it was critical to analyze the availability of labor, along with traditional site location factors. In particular the company wanted a process for easily predicting how much labor is available now and will be available in 10 years in a market. "For us, labor has become a very big concern," said Wayne. "Companies can save millions by analyzing this carefully."

Wayne said that the company was able to create a rapid-response analytic method using PopStats data and SRC technology. "Our old school system used spreadsheets and took three days per model to create. With our new online tools, we built a custom model with location factors important to our company and made it available for everyone to access. We call it Orion, which means location hunter." Among Orion's benefits are that "the entire company is using our system to assess market viability, and any department can research any market whenever they need to gather sight-specific information."

Attendee Feedback:

• Brought up interesting questions about possible applications.

SESSION 4 - The Secret is in the Gaps

Presenter - John LeTourneux, The Kroger Company

Kroger has identified a $120,000 sales opportunity in each of its 113 product categories, which it hopes to access by closing the gaps in lost sales. Kroger currently has $76 billion in sales. "If we could close all of the gaps, we could add another $32 billion," noted John. The national grocery store chain has created an analysis process with the hope of doing just that.

"We have limited resources to invest in remodeling and remerchandising, so we need to strategically plan our spending. We believe this is one key area. John explained this is a massive project and is on the ground floor stages. He said that Kroger is experimenting in one division in Southern California by identifying stores, applying the research, and reviewing results. The steps include:

1. Determining consumer demand in each store
2. Determining the gaps through gravity models
3. Filling the gaps
4. Assessing results

Kroger will report on the results of this analysis in a future STI conference.

Attendee Feedback:

• Very interesting. Would like to see the results in the future.
• Concept was relevant.

SESSION 5 - Extending the Market Research Department's Insights

Presenter - David Hanssen, CVS

David presented an overview of how CVS's GIS department extends demographic data across its 7,000+store enterprise via a web-based system. He said that the company's data is currently used 78% for site selection. But added that there is an effort to make it useful and accessible to other departments, such as marketing, merchandising, and pricing.

Among the motivations, "We want to help provide better data across the company so that all departments can make better decisions," noted David. Also, with a lean five-member GIS team researching 700 sites in 2008, extending demographic data direct access to other departments is "more efficient and takes up less of our time."

To make it acceptable to every department, the data is stored in a secret service warehouse. The system is supported by a vendor that provides data handling tasks, including storing and managing the data, and overseeing web access. Departments are starting to access the data in the following ways:

• Merchandising - to select product mix in individual stores
• Advertising - to create different messages for different stores
• Pricing - to adjust it based on local population demographics
• Attack analysis - for targeting marketing campaigns to specific consumers

Attendee Feedback:

• The CVS presentation was very relevant.
• Great.
• Helpful for the non-GIS professional.
• The CVS portion of this segment was very good.

SESSION 6 - You Buy What You Think

Presenter - Cindy Reid, Mapping Analytics

Cindy explained that companies can maximize the power of their data investments by extending use to departments outside of GIS, such as marketing. Data sharing also prevents common problems such as "marketing talking to different customers than the stores are actually serving."

Cindy said that data sharing is particularly relevant with STI: LandScape's consumer segmentation data, including its latest addition: Mediamark's consumer panel data.

"If you don't use consumer data, you are stuck with only identifying by trade area data," noted Cindy. While entering the arena of consumer segmentation is highly valuable in a variety of ways, she stressed the importance of understanding what you are doing with the data and how you are using it to gain the maximum value.

Highlights of Cindy's talk included:

• Customer segmentation is the study of customer differences, including what they have the propensity to purchase or not purchase.
• LandScape's unique lifestyle indices add another dimension to study customer differences at the block group level beyond the segments, for example, by conservative versus non-conservative consumers living within the segments. "This is a whole lot more powerful than just looking at the customer segments."
• Mediamark data on consumers' purchasing and media habits allows companies to better understand how consumers live and shop in their world. "This data gives you a lot of insight you can't get any other way, so you can better target ideal customers and omit non-customer candidates," noted Cindy.

Attendee Feedback:

• It was great!
• Cindy presents her concepts very clearly.
• Cindy always does a great job!

SESSION 7 - Request for Enhancements

Presenter - Robert Welch, STI

Robert told conference attendees about the opportunity to request specific data they would like to see added to PopStats. "Our approach to product development is to add variables our clients actually need," he noted. "Our philosophy adding new content to our existing products, not just add more new products to our suite."

Allowing customer input that guides product development also "allows us to extend our products in new directions we would not have otherwise thought of." Robert encouraged this year's attendees to "be creative in their data requests," Then he cited two examples of product extensions based on client feedback at past conferences:

• Clients asked for five-year forecasts on incomes in 2007 and that data was added in 2008.
• Clients asked for five-year forecasts on race in 2008 and that data was added in April 2009.

Attendee Feedback:

• This is always great!
• A needed part of the conference.

SESSIONS 1 & 2 - STI: PopStats Methodology

Presenter - Robert Welch, STI

Last, but definitely not least, Robert rounded about the two-and-a-half day conference with a half-day explanation of the PopStats's methodology. This is always the highest-scoring session of the event. Robert said that the PopStats estimates are calculated on four 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's 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+4's 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's 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:

• Very informational and appreciated for the first-time attendee.
• Great stuff! Enjoyed it. I can now answer questions I receive from others at my company with more confidence in the data.
• This was extremely helpful in understanding the data and how it is created. Will be very useful information to know.

Announcements