STI: PopStats 2015 Research Conference Report

This Report provides an overview of the 2015 STI: PopStats Research Conference, including synopses of all sessions. It provides a review for this year's attendees, as well as for PopStats users who could not attend this year.

This year the STI: PopStats Research Conference celebrated its 10th year in style at the Renaissance Hotel in Austin's northwest Arboretum area. Founder and creator, Robert Welch, kicked off this year's event with a look back at its origin. Over 100 PopStats users, vendor partners, and companies considering using PopStats attended the 2015 PopStats Conference. Several more would have attended were it not for the harsh winter weather ravaging the U.S. In fact, considering the widespread weather issues battering the country, it's heartwarming to see how many market researchers made the trek to Austin - it is a testament to the value of the PopStats Conference for many of today's leading market researchers.

The Conference's wealth of information shared by PopStats users, partners, and team members continued the PopStats' tradition of being one of the best places to learn about data's impact on a wide range of business decisions - from where to open new stores, to testing new store concepts, to optimizing cannibalization. Check out this quick overview of the 10th Annual PopStats Conference.

Feedback from Conference Attendees!

Every year, the feedback from PopStats Conference attendees helps us understand how we did this year - and how we can do better in future Conferences. We're proud to say that we receive a lot of great feedback this year. Here are just some of the feedback we received from 2015 Conference attendees.

Notable Stats

• Nearly 90% said they were very satisfied (81.48%) or somewhat satisfied with the conference.
• Over 80% of the attendees said they will attend the conference next year.
• 100% of the attendees said they will recommend the conference to their peers.
• 91% of the attendees said they come to the PopStats for the content.

General Comments

• "The speakers are what got me excited about this conference. They are why I decided to attend."
• "I enjoyed hearing about new companies and their research strategies in the industry."
• "I've heard great feedback about your conference in the past. Robert thinks through every variable. I am impressed with the whole STI team. I'm very glad I attended. I appreciate the fact that we paid for a solid conference."
• "Enjoyable and worthwhile conference as always! Thanks."
• "I deal with a lot of intelligent people at my job, but not as many intellectuals as there are here. This is like going to a brain-spa."

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

• Much more think-tanky and valuable.
• Less sales, more content focused. I like that.
• Excellent.
• Much better.
• Above average.
• More content rich than others.
• For learning, much better.
• Better content and speakers overall vs. other conferences.

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

• Knowing and confirming that what I'm doing is on the right/wrong track.
• Boomtown presentation.
• PopStats methodology presentation.
• The amount of detail.
• Networking.

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

• Mobile data.
• Segmentation.
• Indexing.
• Use of segmentation and better use of EVI and other statistical indices.
• Segment at census block group level and demographics at trade area level.
• Taking your output and apportioning it down to smaller geographies.

This Conference Report provides an overview of the 2015 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

Mon. March 2, 2015

SESSION 1 - Art vs. Science: Real World Sales Forecasting at REI, Curt Newsome, REI
SESSION 2 - Driving Customer Engagement: Location Intelligence and Inflection Points in Spending Behavior?, Al Beery, Pitney Bowes
SESSION 3 - Beyond the Sales Forecast: Other Ways to Enhance the Store Portfolio, Devon Wolfe, Tango Management
SESSION 4 - Real Estate Strategy: Complimenting Modeling Capabilities With a Strategic Process for Better Results, Elliot Schiffer, Smashburger
SESSION 5 - Boomtowns: Identifying and Understanding Areas with Fast Growing Industry, Kyle Day, STI

Tue. March 3, 2015

SESSION 1 - Top Growth Markets for 2015, Robert Welch, STI
SESSION 2 - Using Massive Mobile Data Archives to Improve Retail Modeling: Successes and Pitfalls, Laura Schewel, StreetLight Data
SESSION 3 - A Market Planning Approach, David Chipman, Men's Wearhouse
SESSION 4 - Best Practices for Integrating Real Estate Analytics, Bill McClave, Birchwood Resultants
SESSION 5 - Integrating Real Estate Decisioning and Marketing Activation to Power Results, Anders Ekman, Data Mentors
SESSION 6 - Cannibalization and Transfer: A Brick and Mortar Retailer Perspective, Brian Quinn, Tractor Supply
SESSION 7 - PopStats in Action, Robert Welch, STI

Wed. March 4, 2015

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

SESSION 1 - Art vs. Science: Real World Sales Forecasting at REI

Presenter - Curt Newsome, REI

REI rarely closes stores. However, last year it closed one. "If we had the model we have now when he opened that store, we never would have opened it," stated Curt Newsome at the 3015 PopStats Conference. "Our store location history has gone through many phases, including having no strategy whatsoever. Now we're in the `age of enlightenment,' where we open stores based on both the science of analyzing revenue potential and the art of our real-world experience."

In his presentation, Curt explained how he created the company's revenue forecasting model and gave three examples. He explained that the process begins with segmentation using STI: LandScape. "For REI, customer segmentation is the critical first step. Correlation of the LandScape segments to our sales is incredible."

After the customer profile analysis, REI executes several additional analysis including:

• Regional Calibration
• Site Characteristics
• Quadrant Analysis

Attendee Feedback:

• Very good.
• Relatable to the issues we have with store growth.
• Relevant and insightful.

SESSION 2 - Driving Customer Engagement: Location Intelligence and Inflection Points in Spending Behavior?

Presenter - Al Beery, Pitney Bowes

What major life stage events have the biggest impact on people's spending - starting college, new jobs, new babies, marriage, moving, or retirement? Al Beery began his presentation with this question. If you picked moving, you're right! Pitney Bowes has been studying this topic for 10 years and found that moving impacts consumer spending 10 times more than the other five life stage events.

Further, 23.1 million households move each year - creating $300 billion in sales. The greatest percent of sales come in the month before and the two months after tahe move. But elevated purchases continue for between four and 12 months after the move.

Al stated that retail categories matter. Consumers who are moving make purchases for four basic reasons:

• To restock and replenish - such as groceries and home supplies
• To remodel or built - such as home improvement
• To redecorate - such as bedding, towels, and curtains
• To renew relationships - such as restaurants, banks, and healthcare

Attendee Feedback:

• Where can we get more data?

SESSION 3 - Beyond the Sales Forecast: Other Ways to Enhance the Store Portfolio

Presenter - Devon Wolfe, Tango Management

Devon Wolfe of Tango Management made a case for how to execute market analysis that adds value beyond site location - to effectively "break down the common silos within companies." This helps analysts optimize market research as well as "provide move value in their organization," he stated. Devon presented three scenarios of his theory in action.

1. Rapidly Growing QSR. The challenge was to deal with the problem of franchisee territory encroachment. Tango developed a web-based solution that achieved several objectives including: some basic market research for dealmakers in the field, more advanced analytics for desktop users, and a virtual desktop for mobile power users.
2. Apparel Retailer. This company was developing new store prototypes to keep up with market trends. A program was created that analyzed stores and markets to determine - locations for new stores, locations for the new prototypes, stores that needed rebranding, stores that need to be closed, and stores that needed a combination of old and new prototypes. The research produced results for several teams, including real estate, finance, and construction.
3. Brick and Mortar Store plus .com. This business needed to know the impact of its website on its physical stores and visa versa. Tango created a fast master data management tool that could be used for customer, product, and site analysis.

Attendee Feedback:

• Mentioned things that are relatable and can be used to tweak our analysis.

SESSION 4 - Real Estate Strategy: Complimenting Modeling Capabilities With a Strategic Process for Better Results

Presenter - Elliot Schiffer, Smashburger

Elliot Schiffer introduced conference attendees to Smashburger's trade area research process, which started, he says, with a simple spreadsheet and evolved into a multi-layered modeling process that delivers better results for the QSR, which is adding 100 new stores a year.

The six features of the company's market research model include:

1. Market Mapping - This includes three tiers of characteristics: brand building, solid demographics, and infill opportunities.
2. Model Progression - Tracks how markets change over time.
3. Accurate Model Inputs - Allows trade area tweaks for "human factor" issues such as food pod clusters and gut instinct.
4. New Market Decisions - Allows the company to proactively analyze new trade areas for opportunities.
5. Strategic Pipeline Review Using GIS - Creates store maps for meetings with directors.
6. Micro-Market Development - Helps determine how to build out fast in small area.

Attendee Feedback:

• Very effective speaker.
• Good presentation on the tools and technology that he used.
• Loved the openness of the data and process.
• Hit upon points very relevant to building a dependable model and app to serve multi- department needs.

SESSION 5 - Boomtowns: Identifying and Understanding Areas with Fast Growing Industry

Presenter - Kyle Day, STI

Identifying "boomtowns" before anyone else does is the Holy Grail of location analysis. But despite the best analytics, many companies miss the boomtowns that come out of nowhere. In this session, Kyle Day, explores research options to answer two essential questions about boomtowns: (1) Are there data sources that can help companies identify up and coming areas before the competition? (2) If you do identify the next boomtown will it keep growing?

Kyle began with an overview of the three types of boomtowns: (1) Resource Boomtowns, like Tombstone, Arizona, which are dependent on the price of the commodity; (2) Regulatory Boomtowns, like Las Vegas, which are built around local laws and opportunities; and (3) Specific-Industry Boomtowns, like Detroit, which are dependent on the strength of the industry.

Kyle provided an overview of how he analyzed seven markets that were experiencing significant job growth a few years ago using Quarterly Census of Employment and Wages (QCEW) from the Bureau of Labor Statistics. He demonstrated a unique way to model the data to determine if these boomtowns will continue to boom, flat line, or bust.

Attendee Feedback:

• Could tell he knew his stuff.
• Good job.

SESSION 1 - Top Growth Markets for 2015

Presenter - Robert Welch, STI

Robert led 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 nine years ago, because companies can make year-to-year comparisons. Robert overviewed the top growth markets and pointed out major changes from last year - most notably, the move of Austin back to the number 1 spot. Also, NW Las Vegas returned to the top 10 after a long period of slow growth.

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, 27 markets made the Top Growth Market Report. The top 10 are:

1. Austin, TX
2. Fort Meyers, FL
3. Houston, TX
4. S. Marion, FL
5. W. Orlando, FL
6. Charleston, SC
7. Nashville, TN
8. Raleigh, NC
9. DFW, TX
10. NW Las Vegas, NV

Attendee Feedback:

• Always a fav!
• Loved this one.
• This part is always informative and eye opening and in certain cases predictable.
• This presentation seems to be consistent every year, no need to change.
• Maybe a couple of bottoms too - 2 or 5 biggest losers.

SESSION 2 - Using Massive Mobile Data Archives to Improve Retail Modeling: Successes and Pitfalls

Presenter - Laura Schewel, StreetLight Data

The rapidly expanding world of mobile devices is not just expanding communication capabilities, but also creating one of the largest databases of locational data from millions of anonymous mobile phones, connected cars, apps, etc. StreetLight Data calls this "massive mobile data analytics," and uses it to help companies analyze their retail challenges in new ways. Laura Schewel overviewed how this rapidly emerging new analytic field impacts market analysis.

Laura began her presentation by explaining the four primary sources of location data and the strengths and weaknesses of each: cellular, GPS, apps, and beacon. "Consumer behavior has become much more complex, which is creating confusion in trade area research. For example, consumers today are as likely to shop for groceries one mile from where they work on their way home than they are to shop near their home. Mobile data can help track these emerging new behaviors."

Laura also overviewed what companies can expect to see from locational data sources in the next five years:

• Improvements in spatial precision from cellular devices
• Massive increase in GPS
• More data from wifi networks
• Integration of various technologies
• Regulation changes in the U.S.

Attendee Feedback:

• Learned a lot of information about a new and growing topic.
• Fast speaker, but brilliant.
• I've heard of this, but this solidify it for me.

SESSION 3 - A Market Planning Approach

Presenter - David Chipman, Men's Wearhouse

David Chipman's presentation this year was a follow up on his presentation from the 2014 PopStats conference on the company's biggest market research challenges. This presentation was given to Men's Wearhouse's board of directors based on a challenge: How many new stores can the company open, in which markets, and how much revenue can it generate?

"We wanted to find areas where we both lacked market share and where incomes justified opening new stores," said David. "Plus, we wanted to maximize market share while minimizing impacts to existing stores. We took a simple approach: estimating remaining demand for new store growth by market." ?

The market research process included these seven steps:

1. Determine customer spending in trade area
2. Determine customer sales in trade area
3. Determine customer average store sales in trade area
4. Determine customer sales divided by customer spending
5. Determine target market share
6. Determine sales potential
7. Determine market share

SESSION 4 - Best Practices for Integrating Real Estate Analytics

Presenter - Bill McClave, Birchwood Resultants

Birchwood Resultants session included four speakers: Bill McClave and his partner, Lissy Bethmann, Melissa Dean from La Madeleine Cafe, and Mike Simon from Tetrad. Together they presented a case for identifying the best customers, trade areas, store sites, and opportunities for sales optimization. Birchwood Resultants works with chain restaurants and retail companies employing its custom-built real estate modeling and unit-level performance disciplines.

Trade area must-have's, said Bill, include "alignment with consumer profiles and sufficient density." Lissy added, "Understanding LandScape data is absolutely critical - it's our first building block. In fact, it's so important that if a company won't use it, we won't work with them. When you look at correlation, everything pales in comparison to consumer lifestyles."

Melissa Tyne discussed how La Madeleine searched for new real estate for its new store prototype. Mike overviewed what his company's software adds to the market research process. "We're the glue that holds together the real estate process and the analytics." He provided examples from a QSR, fitness club, and apparel retailer.

SESSION 5 - Integrating Real Estate Decisioning and Marketing Activation to Power Results

Presenter - Anders Ekman, Data Mentors

Data Mentors is a data quality, data management, and business intelligence solutions provider, whose goal is to "help clients make money with data," said Anders Ekman in his presentation. "There's a revolution underway in the world of data with new kinds being created every day. We look for the competitive advantages in it all."

Data Mentors works with something it calls Data as a Service, or DaaS. "We define DaaS as `getting money in the door.' We want to help our clients find that one piece of data that would make the most impactful financial decisions for them."

Anders said that the company's approach to data is a three-tiered process:

1. Foundational data - Focus on building a better long-term data set.
2. Onboarded data - The science of connecting offline and online data sources.
3. Fast data - Augmenting core data set with all kinds of other non-traditional data.

SESSION 6 - Cannibalization and Transfer: A Brick and Mortar Retailer Perspective

Presenter - Brian Quinn, Tractor Supply

"We will take a risk to open a new store near an existing store if we believe that overall store sales will increase," stated Brian Quinn in his presentation on analyzing store cannibalization. His session this year was a follow-up to his session last year on how Tractor Supply finds its target customers, who are recreational farmers and ranchers.

Brian said that some of the company's 1,405 stores are "in some tight trade areas. Some stores are just 10 miles from each other. If our ideal customers are there and the net return is there and we're not negatively impacting existing stores, we'll open a new one."

To approach store growth with this attitude requires rigorous analysis. "However, calculating cannibalization is one of the easiest numbers to get wrong," noted Brian. To get it right, Tractor Supply uses a multi-part, in-depth analytic modeling process to assess each new location under consideration including: identifying new sites, conducting initial predictive analysis, conducting impact adjustment analysis (including sister store, cannibalization, and competition), analogue modeling, and sales forecasting.

Attendee Feedback:

• Great topic.
• Interesting segment to consider.
• Interesting aspect of how cannibalization behaves in rural areas.

SESSION 7 - PopStats in Action

Presenter - Robert Welch, STI

Robert demonstrated a variety of research techniques and PopStats data fields that are included with every release of PopStats, but are somehow overlooked by many users, including the following:

1. Designing a Trade Area - Robert illustrated a method for designing a trade area without the benefit of customer spottings and with minimal knowledge of a store unit's draw to its surrounding community.
2. Growth Analysis - Robert compared a traditional growth map that shows trade area growth based on past data to a map that shows the trade area's growth phase, such as linear, exponential, and plateauing growth.
3. Forecast Stability - He described how to attain the growth forecast spread of a trade area, including three variables: its estimated growth average, the high bound of its estimated growth, and the low bound of its estimated growth.
4. Wealth Economic Data - He showed how PopStats measures wealth based on three data sources: income diversity, average wealth, and poverty.
5. Housing Economic Data - Robert overviewed how mortgage data is used to determine housing price risks.
6. Vitality Economic Data - Robert showed how to measure a trade area's economic vitality based on unemployment rates, a salary index, and occupational demand data. People who feel good about their jobs have a positive attitude about the economy and spend more.

Attendee Feedback:

• I like seeing the behind the scenes stuff.
• Great session.

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