How to Use Data to Plan a Targeted Library Campaign


Use_Data_Targeted_Library_CampaignBig data is heavily used in marketing and sales in a wide range of industries, and it’s also used by political campaigns to target messaging to voters. This data includes information about consumer buying habits, socio-economic indicators, gender, race, preferences, geolocations and much more. Each of these points of data can be put together to paint a broad picture of the consumer and help companies or organizations understand the needs of each individual.

Private sector marketers hire companies that provide data points on consumer habits and demographics to categorize large groups of people into smaller segments. This gives them a comprehensive view of their customers’ and potential customers’ buying habits and preferences, allowing them to segment and target specific groups with the type of messaging they are most likely to respond to.

A well-targeted message to the right person can prompt them to take action, such as buying a product, supporting a cause or using a good or service.

Political campaigners use data service providers as well to segment their audience. This allows them to send targeted messaging to potential voters to influence their voting behavior. For example, parents might be targeted with messaging that focuses on the politician’s stance on education, while an older demographic segment might receive messaging focused on health care. It is this type of targeted messaging that resonates with supporters on a more individual level. It can be what truly drives them to become a stronger supporter and to vote for the candidate.

This same model can be used for library marketing, advocacy and activism. Having access to big data and using it to target specific groups of people allows libraries to build larger audiences of supporters and, in turn, influence voters and local politicians who hold the purse strings to a well-funded library.

But how can libraries use data and test messaging on much smaller budgets?

Putting Big Data to Work for Libraries

At EveryLibrary, we build datasets and use targeted messaging to rally support for libraries using a fairly small budget. There are a few great sources of data about supporters, and, undeniably, one of the biggest sources is Facebook. We use the Audience Insights tool on Facebook for Business to identify trends in the group of people who have already identified themselves as people who “like” libraries. Using this tool, we can also identify the audience’s ages and genders, consumer trends, households, purchase habits and top-liked pages.

These data points allow us to create ads that target the most active and supportive groups of Americans with messaging about libraries designed just for them. This model is based on the Ladder of Engagement theory of activism. In this model people lie somewhere on a range (ladder), and the higher they are to the top of the ladder, the more likely they are to take the deepest actions, such as donating or volunteering. The lower they are on the ladder, the less likely they are to care about the cause when they are first engaged. That means by targeting groups of people who are already high on the ladder, we are able to yield much higher results and spend less money than if we paid to send our message out to everyone.

For example, we know that rural parents “like” libraries at a high rate. So we can begin to test messages that rally them to support libraries or take action on behalf of libraries by signing a petition, making a pledge or donating.

Targeting our messaging also saves our campaigns time and money. For example, it wouldn’t make sense to spend significant money targeting audiences like “Metro Mix,” who are people in their mid 40s who live in mega markets and have middle incomes. We know from Facebook data that there are few individuals in that demographic that are already supportive of libraries, and we would yield very low results if we made efforts to engage with them and earn their support.

How to Spend Your Budget on the Most Effective Messaging

In order to build effective messages for library supporters or potential supporters, we run testing by using dark ads on Facebook that are directed only to people in that audience. A dark ad on Facebook is an ad that does not appear on your organization’s news feed, but your targeted audience will see it as a sponsored post. The targeting can be so precise that it’s possible to target a single individual and nobody else on Facebook will see that ad unless that individual shares it.

Once we’ve identified our audience, we are able to start A/B testing our Facebook messages. We do this by drafting a number of different messages that we believe will resonate with these groups with only slight changes to each. We then spend a small amount of money to promote each ad. We often find that one message works extremely well with one audience while that same message will get almost no engagement with another audience. We learn which audiences respond well to specific types of messages. By knowing this, we are able to take the messages that had the most engagement and put more money behind promoting them to the highly responsive audiences. Over time, this kind of testing and targeting greatly reduces costs and drastically increases the number of people who take action.

Once we know what kind of messaging resonates best with each group, we can use that same messaging anytime we target those same audiences on any medium as long as we have the data to build the groups. While larger data building and communication tools are available, and huge data sets are available if you can afford them, for many libraries, they are out of reach. For example, at EveryLibrary we use NationBuilder as our data building and communication tool, and we use a wide variety of data companies to append datasets to databases like Accura-Append and L2 Political. But we also use some data tools that are freely available at many libraries, such as AtoZ Databases and Reference USA. Both of these tools can allow a library to look at dozens of data points of their users and build some interesting models outside of Facebook.

Once a library has those data points and has tested their messaging, they simply need to decide on a communication strategy to reach those audiences, such as direct mail, email or other paid media outlets. By working through this process, libraries will be able to dramatically lower advertising costs; increase engagement, support, and library use; and hopefully encourage their communities to take action to support them.


Patrick Sweeney

Patrick Sweeney

Patrick "P.C." Sweeney is a tireless and innovative advocate for libraries. A 2007 graduate of the San Jose School of Library and Information Sciences, P.C. is a former Administrative Librarian of the Sunnyvale (CA) Public Library and Executive Director of EveryLibrary California, a statewide ballot committee to support library propositions in California. He is currently the Political Director for EveryLibrary, the nation's first and only political action committee for libraries. He blogs at and he is a sought-after speaker and presenter, as well as a lecturer on politics and libraries at the San Jose School of Information Science.