Tag Archive for 'Demographic Data'

Census Changes in Canada Will Jeopardize Data Quality

Published on July 22, 2010 in Tom Exter. 0 Comments Tags: , , , , , , , .

Tom Exter, Ph.D., Chief Demographer, Pitney Bowes Business Insight

Recently, the Canadian government announced its decision to eliminate the traditional long-form Census questionnaire with a voluntary National Household Survey (NHS) in conjunction with Census 2011. While some supporters in the government agree with the change, the news has garnered backlash from demographers, geographers, statisticians and much of the population, including Canada’s Chief Statistician, who has resigned his post due to the Minister’s decision.
 

As a professional demographer with Pitney Bowes Business Insight in Toronto, I have used Canadian census results for the past 12 years, especially those generated by the long-form census questionnaire. Without the long-form census sample, valuable information used in both the public and private sector will be lost. In addition to the arguments for reinstating the long-form census presented by many Canadian organizations and professional societies including the Canadian Population Society, I would like to contribute the following considerations:

  • A voluntary survey, such as the proposed National Household Survey, would not be a sufficient alternative to the mandatory census sample survey. The traditional one-in-five household sample provides good information for every neighbourhood in Canada. In contrast, information from a voluntary sample survey would be biased, even at the provincial and national level.
  • A voluntary sample survey would have a much lower response rate, relative to the mandatory long-form census, and those who do respond would be, by definition, self-selected. Using information from a self-selected sample of unknown and really unknowable bias in health care planning, for example, would have adverse impacts on health care delivery in Canada.
  •  Filling out the long form may be onerous, but it is not an “invasion of privacy.” The rigorous confidentiality standards of Statistics Canada actually protect the privacy of Canadians because the individual responses are highly protected and only used in privacy-friendly ways (aggregated to relatively large geographic boundaries, for example) to generate information for businesses and government agencies.

Overall, the long-form census data are a significant contributor to the Canadian economy in both the private and public sectors. Businesses rely on census information to grow and help their customer base. Government agencies plan the delivery of services and the allocation of funds to government programs. The quality and utility of the long-form census data are also a testament to the highly professional staff at Statistics Canada who collect, compile, analyze, and disseminate the data to businesses and communities alike.

The significance of this decision for all users of Canadian demographic data cannot be overstated. Readers are encouraged to voice their concerns directly by writing to:
The Honourable Tony Clement
Minister of Industry
House of Commons
Ottawa, ON
K1A OA6

PBBI Canada is interested in your perspectives and questions as well. Please address them to tom.exter@pb.com.

In the Press: Schoolcraft College Taking Strategic Approach to Targeting Potential Students

Published on March 29, 2010 in Leslie Nogue. 0 Comments Tags: , , , , , , , .

Leslie Nogue, Pitney Bowes Business Insight

In the Spring 2010 edition of Community College Technology Update, Marty Heador of Schoolcraft College discusses the value of integrating consumer segmentation data and analyses into direct mail programs.   Schoolcraft College, a continuing education college in Livonia, MI, reached out to the predictive analytic consultants at Pitney Bowes Business Insight for help in maximizing the value of their direct mail campaigns. 

Through effective customer profiling, PBBI consultants were able to identify the affect that distance has on Schoolcraft’s student response rates.  They also determined the student count was dropping and the course load was decreasing per student faster than the number of students enrolled.  By uncovering the most “in-profile” students and the “out-of-profile” students, Schoolcraft’s marketing department is now able to re-direct their direct mail towards the carrier routes within the 10 miles of the college with the highest concentrations of potential students.  This effort has allowed Schoolcraft to “gain a deeper understanding” of their target student, but has also helped focus their marketing funds on those areas with the highest ROI. Read full article…

Spotlight on the Customer: Schoolcraft College

Published on March 17, 2010 in Leslie Nogue. 0 Comments Tags: , , , , , , , .

Leslie Nogue, Pitney Bowes Business Insight

The Challenge
Schoolcraft College, a comprehensive, open door, community-based College in Livonia, Michigan was seeking to target their mass mailings to neighborhoods with the highest concentrations of potential students.

The Solution
Through in-depth analysis of customer, geographic, and demographic factors, analysts at Pitney Bowes Business Insight were able to weed out under-producing carrier routes, allowing Schoolcraft College to focus their energies on the neighborhoods most likely to produce students. As a result of the segmentation analysis, Schoolcraft College gained a deeper understanding of its target student, enabling them to communicate with with prospects more effectively.  Using Pitney Bowes Business Insight’s comprehensive customer segmentation analysis, Schoolcraft College was able to identify concentrations of in-profile students, ultimately boosting enrollment and reclaiming valuable marketing dollars through targeted mailings.

To find out how Schoolcraft was able to quantify and utlimately increase their ROI, download the case study.

Keeping Predictive Models Current: Dealing with Continuous Change…Continuously

Published on March 4, 2010 in Nat Evans. 0 Comments Tags: , , , , , , , .

by Nat Evans, Pitney Bowes Business Insight

Most contemporary predictive models, which forecast performance such as sales, customer visits, membership levels, etc., are based on historical data that create “snapshots in time,” using whatever relevant sources were current at the time of analysis. Examples include POS distributions, store and competitive locations, store sales performance and demographic data. But we know operations and the environment changes as soon as a model is completed and put into use. As a result, model accuracy erodes with each passing day as the data inputs into the model or the benchmarks upon which expected performance are based become stale. To be sure, most site selection professionals and researchers attempt to make sure models are as fresh as possible, updating these data elements on a regular and recurring basis. During recent engagements with several long time clients, we have been asked if there was a way to take into consideration dynamic time series data elements to help with forecasting and minimizing risks.

What do we mean by dynamic data?

Many factors may play pivotal roles in retail forecasting and market prioritization. Depending on the level of aggregation, the obvious thought is that a researcher may be able to affect a change in market conditions or individual sales estimates, depending on the application. Indeed, they can significantly sway analyses enough to change even the simplest of decisions, either minimizing risks (if used appropriately) or increasing a company’s vulnerabilities, especially given the current macro-economic climate.
A couple of sources of dynamic data within the context of a static model may include:

• Macro-economic data such as housing starts, CPI (consumer price indices), funds rates, and unemployment percentages either nationally or at varying levels of macro geography – state, county, or CBSA. Such measures provide a look into the health of consumers’ collective behavior, and depending on how the analysis is structured, whether these factors will be leading or lagging indicators of retail growth and consumer spending (PBBI has created an approach-MarketPulse-that incorporates these factors into predictive models).

• Gas prices. Gas price fluctuations on a regional or even local level can create a similar effect that macro-economic variables may produce in models. Obviously, the higher gas prices rise, the less disposable income consumers will have to purchase goods and services, potentially depressing actual local store performance. Distance may become a stronger deterrent to patronage as a result.

If a retailer’s or restaurant’s sales forecast model was created in better times, it may produce a “false positive,” inappropriately triggering a go/no-go decision and costing company valuable resources and capital from other locations that may be more profitable. Just as importantly, if a company is judging a general or district manager on existing location(s) sales performance based on a projection created earlier in the fiscal year, the company may be unduly influencing that leader’s performance rating on factors outside of his or her control.

How can we create more flexible models using dynamic data?

There exists a myriad of ways we can leverage dynamic data through any forecasting or analytical process, more generally. The important point with any data source is to leverage any and all relationships that may prove fruitful through the forecasting process. But, it must be relevant to your research design, have purpose, and be significant enough to warrant using in modeling and analytical review.

In the future, the ability to collect and cleanse data continuously not only from existing, well-documented sources, but also new sources, such as e-commerce and online social/behavioral data, will become more available and increasingly important across any organization. Additionally, whether on-premise or in the “Cloud”, the technology that facilitates a seamless data flow into predictive applications should enable decision-making with the most up-to-date analysis possible.

Increased Store-Brand Purchasing and the Importance of Store Localization in a Down Economy

Published on November 5, 2009 in Nat Evans. 0 Comments Tags: , , , , , .

Nat Evans, Pitney Bowes Business Insight

A recent AP article seems to underscore the substantial shift in consumer sentiment toward less expensive goods and services that has been going since even before the official start of the current recession. Consumers at all ends of the economic spectrum are much more willing to preserve cash and give cheaper store brands a try, which may be an opportunity for Kroger, Safeway and other purveyors of in-house manufactured goods to gain market share in customer segments once not thought possible.

The article got me thinking about if and how these retail grocery chains, or other retailers for that matter, differentiate their product between customer segments. If the sale of store-branded grocery products has gone up 10 to 15% nationally in the past year, is that percentage equally distributed among all stores? The answer is most certainly not. Does it mean that all categories of merchandise will differ? Not necessarily. Only the data can tell. 

Regardless, consumer preferences are still going to be a driving force, even in a down economy.  It is critical, especially in these times, for a retailer to have the ability to micro-merchandise to the primary customer segments that lie within any store’s primary area of influence (or trade area). Store locations for any chain or consumer packaged good are bound to have significantly different demographic and psychographic characteristics with each trade area, and as such, merchandising the stores in an identical fashion limits optimal market penetration. By examining merchandise shopping patterns and analytically “clustering” stores that have similar characteristics, a retailer will be able to identify trends and compare store groupings of like characteristics.

The information derived from store clustering analysis and localization will be increasingly pivotal for local store planning, marketing, merchandising, cross-promotional activities, and in general, maximizing business potential. This type of analysis is fundamentally important and should be maintained as an integral part of any organization’s regular analytical program. 

For additional thoughts from our Predictive Analytics Consultants, download a free whitepaper on this subject.

Customer Segmentation: Canadian Style

Published on November 3, 2009 in Sebastien Rancourt. 2 Comments Tags: , , , , , .

Sebastien Rancourt, Pitney Bowes Business Insight

Canadian privacy laws set ground rules on how organizations may collect, use and disclose personal information. Under the Personal Information Protection and Electronic Documents Act, for example, personal information can only be collected when it is gathered with the knowledge and consent of the consumer—and only used for the reasons for which it was gathered.

Despite these data challenges, marketers and strategic planners have found effective ways to understand customer needs and create actionable customer segments. These insights and best practices—while particularly germane in Canada—are relevant to anyone looking to improve results by targeting more effectively.

Today’s leading solutions begin with geo-demographic clusters. While cluster segmentation strategies have existed for decades, contemporary clustering methods use robust statistical data and advanced analytical power to capture, create and measure more precise customer segments based on geography, demographics and lifestyles. With the right data and analytical tools, organizations can characterize the behavior of every clustered customer—from their favorite movies and foods to their preferred attire and avocations—enabling users to more accurately predict customers’ responses to every campaign.

Professionals in retail, financial services, media planning, real estate and restaurants, among others, rely on cluster segmentation to improve decision making and business results. Yet with the enhancements made in recent years, some marketers have yet to incorporate the latest advances which can boost overall performance. In speaking with experts across Canada, we’ve identified a series of best practices to help guide your next steps.

Segment by neighborhood, not postal codes. Some segmentation strategies rely on postal codes, which can lead to problems down the road. Each month, as many as 5% of the roughly 850,000 six-digit Canadian postal codes change, as Canada Post updates this system solely on the basis of their mail delivery needs. Not only does this taint campaigns in the short-term, it makes it nearly impossible to manage year-over-year modeling and analysis.

The best neighborhood segmentation clusters begin with census data at the dissemination area levels—which are the lowest levels for which reliable census data are published—providing hundreds of reliable data variables. In addition to data accuracy, these neighborhood-based models offer year-over-year consistency, so marketers can build on past success over time.

Incorporate household-level insights. This past year, leading cluster models have found ways to use more comprehensive household level data, incorporating consumer information that goes far beyond census findings. These inputs, which conform to Canadian privacy laws, represent an unprecedented level of detail and behavior-based data—and create a more high-definition view of customers and prospects.

Maximize data points. Not all household level data is the same. Some cluster models are built extrapolating data from as few as 8,000 surveys across the full population of 33 million Canadians. More reliable cluster models will analyze self-reported data from as many as 10 million individuals—providing for more accurate targeting and a lot less guesswork.

Overall, organizations that employ these best practices will benefit from a multidimensional framework that makes it possible to sort through the complexity of Canadian consumer culture without having to manipulate literally hundreds of census and survey variables.

One such solution is PSTYE HD, the Pitney Bowes Business Insight segmentation system created using an innovative two-step clustering process. The 59 clusters identified, including Canadian Elite, Joie de Vivre, Urban Verve and Next Gen Rising, leverage the largest and most robust repository of Canadian consumer intelligence to date—making it easier for organizations to locate new opportunities, connect with customers and communicate more efficiently.

Learn more about PSYTE HD at www.pbinsight.com/psytehd. As always, we look forward to your feedback!

Notes from the ICSC Research Conference 2009 in Phoenix

Published on October 29, 2009 in Devon Wolfe. 0 Comments Tags: , , , , , , , .

Devon Wolfe, Pitney Bowes Business Insight

About 170 researchers and industry professionals gathered in Phoenix for the annual ICSC Research Conference, which is a gathering that has always been part networking, part content. The numbers this year were down considerably from years past, but the group was still spirited and engaged.

The ICSC group has long been dominated by the department store and shopping center research departments, yet this year, the higher numbers of attendees were from value, low-price point retail, just as we’re seeing in the sales results posted by various chains. Drug stores, dollar stores, and discount apparel were all well-represented. The conspicuous absence was big-box specialty retail. Very few attendees came from that segment of the industry, likely due to the slowdown in large store construction.

Instead, many operators I talked with are opportunistic and looking for great deals in the marketplace, while the developers and shopping center owners are hoping that the coming commercial mortgage-backed securities (CMBS) storm doesn’t wreck the rest of their business. Economists presenting at the conference were quick to point out that while we’re nowhere near recovery at this point, it’s inevitable that things will start to pick up within the next year, but slowly. Even though we want to think that this recession is drastically different than all others in the past, it isn’t necessarily. In the past, just as today, job recovery tends to follow market recovery, which of course means that it’s going to take a while for retail spending to recover completely.

On the methodology front, one thing to watch and prepare for is the 2010 U.S. Census, which has the distinction of being the first where the American Community Survey (ACS) will replace the long form in its entirety. Without listing all the details here, the important thing to remember about the ACS is that it uses sampling gathered on a periodic basis at different levels of geography. This means that while state level information will be reported annually for the previous year, block group information is reported each year for an average of the previous 5 years’ surveys. Sound confusing? It will be. We at PBBI are working on solutions to help take the guesswork out of using these data. Stay tuned . . .

In the meantime, we welcome you to download a free whitepaper on the impact the ACS can have on your business.  We also encourage you to visit the census website for more information on the ACS.

TransPromo: Drive Results with Behavioral Marketing

Published on October 19, 2009 in Leslie Nogue. 0 Comments Tags: , , , , .

Need a powerful marketing tool that enhances operational efficiencies? Need to increase revenues and generate a stronger return on your marketing investment? Leveraging customer behavioral data to drive TransPromo, offers the ability to increase campaign effectiveness while adding value to the customer relationship.

So, join us for a free webinar on Monday, October 26, 2009 at 1:00 pm EDT and learn more about transforming your transactional statement into a powerful marketing channel. This session will help you identify which customer information to leverage in your statements and how to append data to increase relevance of the offer. Also included in this presentation are pertinent examples of the creative and strategic opportunities available to marketers. Register now

Recommendations from PBBI’s Predictive Analytic Practice Leaders

Published on September 4, 2009 in Leslie Nogue. 0 Comments Tags: , , , , , , , , .

The July edition of Response Magazine features an article by Al Beery and Brian Hill, Practice Leaders in Predictive Analytics for Pitney Bowes Business Insight, on the importance of companies gaining a better understanding of their target customers to enhance their marketing campaigns. The resulting article highlights the benefits of location intelligence through the use of demographic, psychographic and macroeconomic data to help companies make smarter, more strategic decisions.

For more on the recommendations from our Predictive Analytic experts, visit Response Magazine.

Response Magazine is a monthly publication geared toward professionals involved in all facets of direct response marketing (circulation: 18,626)

When two worlds collide

Published on September 3, 2009 in Steve Seabury. 1 Comment Tags: , , , , .

Steve Seabury, Pitney Bowes Business Insight

While the 2010 U.S. Census is still months away, a recent advance in data analytics demonstrates how amazing things can happen when customer and location intelligence comes together.

For years, real estate specialists and strategic planners have relied on spatial analysis to make decisions that required significant investments. The power of location intelligence proved invaluable on many fronts. The stability of neighborhood demographics enabled decision-makers to hone in on trends that could impact long-term profitability. The precise nature of geocoding provided for year-over-year consistency. Plus, the ability to visualize and map customers, prospects and competition against existing and planned sites led to key insights… insights that have enabled banks, retailers, utilities and many other industry executives to exceed expectations.

At the other end of the spectrum, marketers turned to household segmentation models. Robust demographic data at the household level could be used to create clusters—segments of consumers who shared similar lifestyles, characteristics and needs. This lifecycle approach made it easy to target the ‘retired affluent’, ‘young families’, ‘single post-grads’ and dozens of other key markets. And with records updated quarterly (or even more frequently), marketers could respond quickly to life events.

Now for the first time, these distinct approaches have been combined to deliver enhanced network performance management and customer analytics solutions. Using deeper, more precise demographic data, organizations can make more informed and timely decisions about critical real estate and marketing initiatives. These next generation demographic data tools incorporate advantages from both disciplines and can help organizations overcome today’s top challenges, for example:

  • Enables marketers and strategic planers to work from the same platform
  • Compares changes in household make-up with neighborhood shifts to uncover pockets of opportunity
  • Normalizes household data to block out the noise of short-term events to create more accurate projections
  • Eliminates the need for ZIP Code targeting, which rarely reflect true neighborhood and lifecycle segments
  • Links store network performance with customer relationship management strategies

Of course, creating the best of both worlds requires you to start with the best in both worlds. That’s why Pitney Bowes Business Insight teamed up with the Gadberry Group and Acxiom® Corporation and their PersonicX® segmentation system.

These data sources compile consumer data from over 100 sources, including public records, the U.S. Census and self-reported data. Measurements for accuracy and completeness are part of a sophisticated multi-source build process where individual data attributes are compared across multiple providers. While mapping and analytic tools previously dealt with neighborhood and block-level data, these new tools drill down to race, ethnicity, gender, education, marital status, occupation, income and lifecycle on an individual household level.

In many ways, incorporating Gadberry and Acxiom data into PBBI predictive analytics models will enable organizations to bridge the gap between real estate decision-making and marketing strategy – incorporating the best of both.

For more information on the newest technologies, visit us at http://go.pbinsight.com/household-derived-demographics.