A new report published March 29, “Core Customer Intelligence: Public Library Reach, Relevance, and Resilience,” brings together market segmentation from ten public library systems across the United States to explore how libraries can examine and act on granular data about their core customers—the 20 percent of cardholders who check out the most physical materials. Using 2014 customer and checkout data to group top library users by lifestyles, interests, preferences, and behaviors, the study, which was funded by the Institute of Museum and Library Services (IMLS) as part of the 2014 Leadership Grant Program, drills down into community demographics to reveal that core customers aren’t found in any one segment of the population but occur across all lines, reflecting the diversity of their communities.
The libraries—Anythink Libraries, CO; Brown County Library, WI; Denver Public Library (DPL), CO; Houston Public Library, TX; King County Library System, WA; Las Vegas–Clark County Library District (LVCCLD), NV; Omaha Public Library, NE; Pierce County Library System (PCLS), WA; Skokie Public Library, IL; and Topeka & Shawnee County Public Library (TSCPL), KS—represent a mix of large and medium-sized systems comprised of urban, suburban, exurban, and rural residents. Taken together, the libraries serve 7.8 million people, four million of them cardholders who made 67.4 million book and media checkouts in 2014. All participating libraries are clients of CommunityConnect, a software data analysis service developed and managed by CIVICTechnologies, which provides geographic information system (GIS)–based data for libraries and which provided the numbers for the report.
TAKING A CHAPTER FROM CORPORATE
Market segmentation data, which divides a target market into subcategories based on geography and financial, family, educational, or consumer demographics in order to discern customer patterns, has long been a tool of the corporate and retail spheres. While libraries have often relied on national benchmarks and opt-in customer research to gauge their effectiveness, market segmentation allows them to take a hyperlocal look at all corners of their communities in order to answer questions about which parts of the population they’re serving, how they can maintain existing relationships, and what they can do to build new ones.
This is far from the first time libraries have used market segmentation data—Gina Millsap, CEO of TSCPL (and a 2007 LJ Mover & Shaker) has been using it since she worked with the Ames Public Library, IA, in 2000. However, it has since been modernized and customized for library markets. Without the sharp focus that market segmentation provides, said Danielle Patrick Milam, development director of LVCCLD and Foundation and one of the report’s coauthors, most libraries are working with data at a zip code, jurisdiction, or school district level, as opposed to the far more detailed census block group. Added Marc Futterman, president and CEO of CIVICTechnologies, the report’s other coauthor, “We used to draw circles around our branches [on maps]!”
CIVICTechnologies collected a vast amount of library cardholder data over a decade (with privacy safeguards in place to prevent tracking of individual checkouts), and thought it was ripe for the type of analysis usually reserved for business big data. In addition to LVCCLD, which had conceived the study with CIVICTechnologies, Futterman identified nine other libraries among the company’s customer base, all of which, when asked to participate, were enthusiastic about taking their data further. Futterman held several workshops around the project, bringing together administration, branch leaders, and outreach teams from the participant libraries, as well as staff from other departments not usually involved in the outreach process such as marketing, information technology, and human resources.
Once customer and checkout information was collected from CommunityConnect, it was joined to two commercial geo-demographic market segmentation systems developed by LandScape, created by Synergos Technologies, Inc. (STI), and Tapestry, from Esri. Grouping data by census block level produced detailed demographic maps of each library’s service area. Fuzzy Logic, a data analysis firm, performed quantitative analysis of each library’s data. The results of the report were both affirming and surprising.
In the Core Customers study, library districts are grouped by several demographic factors into categories and then segments, each assigned a descriptive name by LandScape and Tapestry. The Industrious Urban Fringe segment, for instance, which is part of the Global Roots category, is characterized: “Family and multigenerational households in single family units are dominant in this mostly Hispanic segment with income earned through employment in the manufacturing, construction, retail, and service sectors.” Urban Squires, from the Crème de la Crème category, are “highly educated, 30-somethings living in urban neighborhoods.” Some 74 highly specific segments are named.
The data revealed a range of patterns. Some metro areas, such as those served by LVCCLD, show diverse segments scattered across the library’s service area; they were surprised, said Milam, at how the demographics shifted block by block. As staff examined the many different household types served by a single branch, she said, “that branch manager had no idea that they were managing basically 21 different kinds of audiences.” Others, such as the Houston metro area, show clusters of similar household segments.
Population trends were also evident. For example, in LVCCLD 70 percent of the cardholding population is made up of families with young children—26 different kinds were identified. These families were divided into three types: higher-resourced families, who are young, educated, and on their way up; lower-resourced families with lower incomes, working predominantly hospitality, construction, and retail jobs, many without a high school education; and families that are low-resourced and are also New Americans, often with English language issues. In DPL’s service area, on the other hand, 66 percent of cardholders are from single-adult households—half of these well-resourced, educated, and working white collar jobs; half younger individuals who are still struggling.
Perhaps unsurprisingly, high wealth households—where income and education levels exceed the national average, often located in an area’s suburbs or urban fringe—also make up a large segment of several of the libraries’ core customer group, having grown by more than two percent annually since 2000 and remaining one of the least diverse segment groups. One of the study’s more unexpected results was the number of single adult households that are core customers—both middle-class and struggling single-parent households.
Latino households are one of the fastest-growing population groups in the nation, and show similar growth among library users. Four of six market segments representing predominantly Hispanic American neighborhoods appear in the top ten core customer segments. These are mainly comprised of family households, mostly with two parents in their 30s, but include younger working families and single-parent households. The study revealed a strong opportunity to grow customer loyalty in the Hispanic American segments, many of whom already hold library cards and can be targeted by services, marketing, and outreach. New Americans also show up on the core customer radar.
POWER USER PATTERNS
Most important, however, was the finding that libraries’ core customers are distributed across the population as a whole, and the top ten market segments for library use were also nearly the same, in all districts, as the top ten segments for core customers specifically—so libraries need to distribute their efforts to gain and retain core customers across all demographics. As the report’s Executive Summary states, “Overall, this study finds that core customer characteristics and behaviors are not homogeneous. There is no simple model of local sets of core customers nationally. Instead, across all ten of the libraries we found that core customers are complex and unique, distributed in distinctive patterns that reflect the complex and unique characteristics of their respective communities.”
This kind of data allows libraries to move away from a one-size-fits-all approach toward more effective outreach and strategies; the report calls for libraries to use the information to improve their reach, relevance, and resilience. Learning about their audiences in such detail, participating libraries are able to gather information about all of their functions: program development, collections, and different focuses in the branches as opposed to out in the community. This, in turn, allows for efficiencies in strategies of service, marketing, and collection development.
The fact that there are power users in each market segment, noted PCLS executive director (and 2010 LJ Mover & Shaker) Georgia Lomax, helps combat perceptions about who uses libraries—not just children, or the poor, or the wealthy. This awareness can be used not only to help libraries plan their next steps, but to make their case to funders and voters.
DATA DRIVEN DECISIONS
The information became a way for library systems to empower their branches to provide a district-wide level of service yet tailor by branch audiences. “In a field where there’s so much change going on and the industry is in disruption,” Milam told LJ, “we need this kind of tool where we can very quickly target the key populations that we are serving or want to serve, and understand at a very deep level what makes them tick and their motivations, their interests, their lifestyles, where they are in their life—information that, quite frankly, we just can’t get if we throw a Town Hall or even if we pay for very expensive focus groups.”
“We were traveling our board around to our various branches,” Milam recalled, “and before the meeting would even start we’d have the local branch manager get up and start with this data.” It empowered both board and staff to say, “we know our communities really well and this is helping us know them even better,” said Milam.
At PCLS, said Lomax, “Our branch people have all taken that [data], learned about what it says about their community, then taken those sheets of paper and a map and driven around their neighborhoods, and looked at what a neighborhood that is a particular segment looks like…. And then come back and worked together to build their annual plan for service, thinking: Knowing this about your community, what one thing could you do differently that you want to try this year to see if it makes a difference?” These analog “windshield tours,” which take cues from indicators like multifamily houses, swing sets, and raised garden beds, don’t always result in big-ticket items, Lomax told LJ, but rather “small, practical things.” One location found that it had a large family-based community with strong faith ties, so it decided to pull and highlight a collection of faith-based fiction and nonfiction, resulting in an uptick of checkouts.
Even single-branch libraries, like TSCPL and Skokie, have much to discover. Before the core users study, TSCPL believed it wasn’t reaching enough inner city households in East Topeka—but in addition to discovering that the library had an almost 90 percent penetration rate into its Inner City Tenants segment, the surprise was that it had very little reach into Green Acres, its outer county markets. “Until we did the initial survey with CivicTechnologies, I think most of us would have said we do a pretty good job of serving out in the county,” Millsap told LJ. “We really work hard to serve people who live far away from the library. But that’s not what the data…told us. Basically the further someone lives away from the library, the less likely they are to have a card.”
Since then the library has repositioned itself in the community, instituting library lockers in outer areas among other services, and focusing collections there on topics that interested those residents, such as health, gardening, and cooking. Millsap has tasked non-managerial staff to each create a service plan around the data, using it to make a case for any new initiative. “You reward staff for doing that; that’s the right thing to do,” she said. “That’s how you get it integrated into your day-to-day practice and thinking.”
In Skokie, the community had shifted during the recession because of housing prices, with former inner city dwellers moving into the community. With a growing international population, the library refocused its bookmobile, developing bilingual materials and specialty programs. The local school districts noticed that children were more ready for kindergarten and partnered with the library on its Little Learners program.
MORE BIG DATA?
Because the data used was tied to customers’ card and checkout data, this study was based on use of physical materials only—books, DVDs, and CDs. But core customers consistently borrow a higher percentage of non-print materials, and the participant libraries hope that at some point a similar study of programming, e-content and downloadables, databases, and Wi-Fi and computers will be possible.
Milam hopes that vendors will be interested in participating as well. “We want the vendor world to get excited about tying to the library cardholder,” she told LJ, “so that we all learn more about who’s using what kind of data as the industry shifts from purely physical to digital, virtual, computer use, program use, even major campaigns like summer reading.” Not only could vendors help bring data in, Milam added, but studies like this could help them understand more about what libraries’ customers want.
Being able to investigate usage of non-physical services will help libraries plan for the future, both in terms of collections and space. “It’s a question that we all struggle with,” said Millsap. “What’s the right mix of digital and print? How should we be planning over the next five years in terms of budget and resource allocation?”
Not only are libraries able to use the data for their own purposes, noted Milam, “We’re able to inform the school district, the city planning office, and other alliances that we’re working with now, around bigger issues of education and economic development, where we think the hot spots in the community are.”
The study has given libraries a more sophisticated view of their communities than they had before, added Lomax. “[The data] helps us do more than just throwing the spaghetti on the wall…. We can make some informed choices.”