STI: PopStats 2011 Research Conference Report

SESSIONS 1 & 2 - STI PopStats Data Methodology, Robert Welch

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

Presenter - Robert Welch, Synergos Technologies Inc.

At this year's New and Major Initiatives discussion, Robert made the following announcements:

• PopStats data's U.S. population estimates in April 2010 beat the U.S. Census Bureau's estimates. PopStats was within 3,422 of the actual Census, announced in December 2010. Meanwhile, the Census Bureau overestimated the population by 232,406.

• Robert introduced a new corporate-level marketing program that includes two action figures, named Xe and Athena. The campaign will officially launch this year. These characters will appear in an ongoing comic-book-style series to promote awareness of the value of data and downside of misinformation across the enterprise.

• With the group's input, a decision was made to release the new 2010 boundary files in October using Tiger files, until other geographic files are ready.

• A new, more robust extraction program will be released in July 2011 for STI data.

Attendee Feedback:

• Very helpful.

• Missed and sorry I did.

SESSION 2 - Top Growth Markets for 2011

Presenter - Max Fisher, STI

One way to understand markets is by looking at them from a historical perspective, said Max Fisher, STI's Colossus Product Manager, during his presentation. He analyzed several markets across the U.S., all culled from the PopStats annual list of the top-growth markets. He applied several data sets within the new Colossus econometric data product to analyze these markets just before the recent "great recession." He found that using the economic data it was possible to have predicted the recession in some markets.

"Population shifts and demographic changes are intertwined with economic forces," noted Max. "Our research found significant differences in the economic health of several of PopStats data's top-growth markets." A white paper was distributed to attendees and is available on the STI website, called: "STI: Colossus Data Predicts the `Great Recession.'"

Attendee Feedback:

• Robert enhanced this presentation with his insights.

• I would like this expanded and the detail on why the markets were succeeding in spite.

SESSION 3 - How to Evaluate New Technology

Presenter - Brian Strickland, Family Dollar; Dale Caldwell, Kroger; Glen Boyer, Ross Stores

Researchers from three companies discussed how they evaluate new technologies before adding them to their operations. Here are some of the many insightful answers shared by the panelists in this session:

Q: Do you have a corporate roadmap for technology adoption?

A: Dale: No. Our research team will review and decide on which technologies we need. We are not just looking for the next new innovation, but for products that have the potential to improve our efficiency, productivity, and accuracy.

Q: When examining new technology what drivers do you consider?

A: Brian: We ask ourselves: Does it add value? Can we optimize it? Can we share the data across the enterprise?

Q: Do you go looking for new technology?

A: Glen: We do. Often new data drives our new technology decisions.

Attendee Feedback:

• This is my favorite session of the day. I really liked the topic and speakers.

SESSION 4 - Developing a GIS Department from the Ground Up

Presenter - Stef Robinson and Phil Weick, Panera Bread

In this session, two members of Panera's GIS Dept. described how they turned around the process of site selection - from a system based mostly on real estate brokers applying their "gut feelings" on "good deal," to one based on site selection data, GIS technology, and an internally developed risk assessment strategy.

The impetus for change occurred in 2002, when the fast-growing restaurant chain was opening more stores, but experiencing lower sales. Once the GIS department ramped up with data and tools the team began finding the problems with their site selection process, such as store cannibalization. Panera created a risk assessment system that ranked store locations from 1 to 5, based on metrics from its highest performing cafes including: population density, retail density, daytime workers, and trade area size.

"The new approach achieved profound results," said Stef. "We reversed the downward trend in sales, and did a complete about face."

Attendee Feedback:

• Totally rocked!

SESSION 5 - Tried-and-True vs. Cutting-Edge?

Presenter - Liz Cruz, Publix; Rob Stewart, Wegmans; Tony West, Whole Foods

In this panel discussion, three researchers from three grocery store chains talked about their company's approach toward market research, including the use of research tools - both traditional and cutting-edge. Here are very short excerpts of each researchers' insights:

Tony West from Whole Foods said: "We use a mix of traditional and innovative technologies to find locations. We use gravity models and other models, as well as lifestyle segmenting. We have a very specific consumer demographic and we don't open that many stores a year, so we are very selective."

Liz Cruz from Publix said: "One of our leading strategies is prioritization. We look for locations that match our specific demographics, which are quite narrow. We don't use gravity models, but are always looking for the next tool to improve site selection, as well as purchasing and loss prevention."

Rob Stewart from Wegmans said: "We will add new tools here and there, but we are not really that cutting edge. We've just started applying segmentation to our gravity miles and are exploring opportunities with that approach."

SESSION 6 - Technology Round-Up

Presenter - Research Vendor Demos: X-Span Results and Trade Area Systems

Two vendors overviewed their product and service offerings for the attendees.

SESSION 1 - Using STI Data for Discovery and Creating Custom Variables

Presenter - Robert Welch, STI

Robert described a variety of simple mathematical and GIS techniques for organizing, filtering, and highlighting data elements that allow researchers to develop new conclusions about their markets:

Expected Value Indicator (EVI). This identifies the kind of growth that is occurring in a market: for example, new growth, old growth, and plateauing growth. EVI acts as a filter to the quality or character of growth within periods of time. This works best at smaller levels of granularity.

Ethnic (Race) Concentrations. This is one of the most popular maps, showing racial concentrations in four categories: Whites, Blacks, Hispanics, and Asians. It provides an informative color-coded map of these concentrations.

(Simpson's) Diversity Indicator. This can be used to visualize concentrations of populations in different categories, such as race and wealth. It doesn't show race breakouts, just diversity levels.

Attendee Feedback:

• Excellent.

• Always helpful to be reminded of the less obvious variables.

• Very helpful to hear Robert talk about custom variables and creative statistical approaches.

• Great side-by-side map comparison and illustrating maps that are complementary.

• Very helpful info that I will apply.

• Great job, need more presentations like this related to practical application of your data and put it into practice. The teaching format is successful.

• My favorite topic!

SESSION 2 Data-Driven Myth Busters

Presenter - Andy Verostek, Del Taco

In an entertaining and insightful presentation, Andy presented a variety of common myths about sub-segments of the population - and busted through those myths using STI data, such as the following:

Myth: Language is the biggest barrier to growth in ethnic neighborhoods. Reality: Researchers can apply PopStats "languages spoken at home" data to determine the percentages of languages actually spoken in homes in a market. They will often be surprised at the outcome.

Myth: Employers equals daytime population. Reality: A market with lots of employers could also have a large population of night-shift workers or employees who work from home. WorkPlace can help breakout the actual estimated daytime employment numbers of a market.

Myth: Cities have personalities. Reality: Income levels are key. High-income levels drive luxury retail and services. Middle incomes drive daily needs and some indulgences. Low incomes drive bare necessities. LandScape helps identify these neighborhood personalities.

Attendee Feedback:

• Love seeing how others use PopStats and other variables to analyze markets and trade areas. Would be great to tap into the minds of other users as dynamic as Andy.

• 5+

• Andy's use of the data was discussed very effectively. Love the addition of the graphics.

• Fun, clear, and interesting.

• Came away with some great brainstorming ideas on things to do in our own "world."

• Entertaining and interesting.

SESSION 3 - Location Intelligence - Going Beyond Demographics

Presenter - Wayne Geary, Jones Lang LaSalle (JLL)

One of the keys to site selection for many companies is labor. They need to first answer the question: "Where is the labor?" before they can ask, "Where should we open up a business?" Wayne presented an over of a technology created at JLL that helps companies find the labor they need to operate their businesses.

The system was created by spending $3 million dollars on labor and staffing. It includes custom-built labor indexes using STI data, including PopStats, WorkPlace, and Colossus. In fact, the econometric data in Colossus is one of the service's key distinguishers.

"We were able to identify multiple geographies of opportunity for our clients, thanks to our robust data populating the system," said Wayne. "We can tell our clients with confidence, `here's the right location for your business.'" The technology won the prestigious Da Vinci Award for Innovation.

Attendee Feedback:

• Understanding the corporate real estate application of demographics is very interesting.

• This is a company I could see us working with, so I found it interesting to learn their data and global coverage.

• Nice to see a top-notch system and data set.

• Very interesting and cool applications of data and technology.

SESSION 4 - Display Holography: Historical Survey and Recent Developments

Presenter - Mark Holzbach, Zebra Imaging

Mark introduced and explained digital holography for the attendees. He provided several interactive examples of the "hyper-realistic" technology in practice. "Having it look just like reality is the goal," he said. "It's being used currently in the military and is being commercialized to provide 3D visualization for complex location-based decision making. It's one area of technology that is growing - and many people are working on it."

Attendee Feedback:

• It sure was fascinating.

• It's nice to see a presentation that is interesting and fun even though not too relevant.

• I found this entertaining and educational.

• Really interesting, but not something I could see us using in a practical setting.

• Interesting, high-tech future outlook was a good topic.

• Cool technology, not very applicable, but interesting none the less.

• It was fascinating and awesome to see someone so enthusiastic about what they do.

SESSION 5 - The Perfect Location - Now What?

Presenter - Susan Cordts, Adaptive Technologies, Inc.

Once a location is selected, expectations for success run high within companies. Everyone wants to see revenue, profits, and customer bases build - and quickly. But sometimes, despite the best intentions and research, a site will underperform. What can a business do in this situation? Susan addressed this issue in her presentation.

She gave an example of a client who had two non-performing new locations. To try to understand the problem, asked: Who are the actual customers? Once the demographics were identified the stores made modifications in product offerings and discounts to appeal to them.

"We help companies better understand the real story of their locations using objective data," noted Susan. "By making business decisions based on that research we turned both of these locations around within 12 months."

SESSION 6 - Assessing Retail Opportunity Using Market Outlook

Presenter - Sam Lowder, ROIC

How does supply and demand in a trade area impact market research? Sam explored this question in his presentation using data from STI: Market Outlook. "You can analyze trade areas on a finer gradient level using supply and demand data," he noted.

He provided several examples of MO data in action. For example, if the data shows that in a particular trade area the demand for auto parts is $1 million dollars of business and there is already a supply of auto parts of $2 million, the trade area would clearly be a high-risk location for another auto parts store. MO provides an opportunity to find opportunity gaps that businesses might otherwise miss.

Attendee Feedback:

• Good presentation and funny too :-).

SESSION 7 - Technology Round-Up

Presenter - Research Vendor Demos: Forum Analytics and Alteryx

Two vendors overviewed their product and service offerings for attendees.

SESSIONS 1 & 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.