How the Brooklyn Public Library used data visualization to build a better library

Jon Reed Profile picture for user jreed February 18, 2015
The Brooklyn Public Library tells the story of their digital journey (and analytics challenges), and how data visualization from Tableau changed their work.

When we think of digital business, a public library might not come to mind. But Brooklyn Public Library definitely qualifies. With an increasing demand for digital content, and a vital role to play in public learning, Brooklyn Public Library faces the same high stakes as any digital company. Their constituents count on them - to the tune of 1 million Brooklynites who carry the library card.


Brooklyn Public Library (or, as the peeps call it, "BPL"), circulated more than 15.5 million books, DVDs, eBooks and other materials in 2014. Computer literacy is vital, supported by 1,400 public computers and free wireless access across all 60 branches. The fifth largest library system in the United States, BPL must deliver relevant content to a multi-cultural demographic in pretty much all stages of life.

To get this right, proper data analysis is essential - that's where Tableau enters the picture. I recently had the chance to talk about the library's digital journey (and analytics challenges) with BPL's Adam Leddy (Communications Coordinator) and Diana Plunkett (Manager of Strategic Initiatives). Here's what I learned.

The library's digital and educational challenge

I asked the BPL team if the library is fading from relevance due to the digital shift away from printed books and periodicals. The answer is a definite no - instead, BPL is proving itself a crucial resource for citizens who want to consume a range of content. Plunkett:

Today, there's more of an understanding of how Brooklyn Public Library is more than just books. More and more people realize that e‑content, including e‑books, but also access to databases and online learning sites, comes with your library card. Then there's the programs and events we hold at the Library. As we do a better and better job of making sure people are aware of all of we do, then we see more and more people coming to the Library.

Leddy pointed out that demand for the overall collection is at an all-time high. Printed books are not going away, as much as they are being supplemented. Given that the digital economy calls for a different set of job skills, I wanted to know how BPL sees the role of the library from a digital literacy and education standpoint. Plunkett:

As the nature of our world changes and the skills people need to be successful in this world evolve, the library, which has always been the place where people come to do their own learning, is just more and more critical. We serve patrons on such a broad spectrum in Brooklyn. There are immigrants who need to learn English. There are professional immigrants who have degrees in their home country, but need to understand how to become professionals here in this country, whether that's through language or through certification.

There are young people who want to explore what different career possibilities are. There are people who are in need of jobs and job training. There are people who need basic skills before they can even begin to do job training, around literacy and digital literacy in particular. There are people who are changing careers and need maybe tech skills to move into a new sector as they evolve and grow as individuals. The spectrum of what we do is just so broad. Each branch needs to understand the needs of their community deeply, and build in the programs that they need.

Leddy added:

We have a very strong online component to our trainings. We recently have started to offer and all the various trainings you can do there, many of which can lead to certifications. It's a lot of professional development.

Better service through better analytics

One of the core BPL analytics needs is the bottom line: reporting requirements. As Plunkett explains:

We report to the state and the city and the government entities who fund us; we have that kind of mandated reporting. Then we have internal reporting where we're tracking what we're doing, so that we can make sure we're meeting the needs of this diverse population. Our work with Tableau has eased all of that.

One key to success? Start on a small scale. Plunkett:

Our IT department recognized the power that Tableau could bring to the library. So they started using it on a smaller scale within their department to capture their metrics. Then when I joined in June of 2012, they already had plans to start moving Tableau into a larger arena.

Of course it takes quite a while to identify where the data is going to come from. The data is always the hardest part of that. We launched a set of metrics that everybody internally in the library can look at in the fall of 2013. We started with our simplest metrics, the ones that were easiest for us to capture. Our data around circulation is pretty clearly defined and pretty clearly understood, so that's where we started. How many books get checked out? How many books get renewed? That kind of thing. That was the low hanging fruit for us.

From anecdotal impressions to data clarity


From those beginnings, BPL expanded their use of Tableau. They now have a system that counts people as they walk into each branch, providing a much more accurate measure of who's coming through the door. They also track program attendance, including location and age of attendees. Public computing use and free WiFi consumption across all branches are also tracked. E-books and physical book use can now be viewed as one set of data.

I asked BPL about the feedback they've gotten to date. Plunkett noted the value of moving beyond the anecdotal, and involving employees in data-informed decisions:

This data is all available to everyone who works at the library so that as we are making decisions, whether they be system‑wide decisions or decisions on a branch level, this data is input to those decisions. It also lends a sense of transparency across the organization, so that everyone can see the factors that were part of making that decision.

Some of the folks in the branches have said, "I didn't realize that this type of data was available!" We can show the checkouts at a particular location are world language books, or juvenile board books, or different types of items. While they knew anecdotally, they've now got really clear data about what types of books are being checked out in their branches, and that's informing both the programs they might hold, and the book displays they feature. They can focus on the material patrons are most likely to be interested in, and make sure that they're easy to find in the branches.

The impact of Tableau's data visualization is another factor. Prior to Tableau, some of this information existed in reports, but getting people to work with the data and make sense of it was another matter entirely:

A lot of the data we are displaying in these visualizations is data that was captured before, but there wasn't an easy mechanism for everyone in the organization to see the result of that captured data all in one place. We find that people are more effective in their reporting because they can see the results. It's not just being reported and it goes into a black hole somewhere. The visualizations make it so that people who aren't used to diving in and mucking with the data can easily take a look at what's going on, and understand what actions they can take as a result of it.

Advice and forward strategy: the future is more ad-hoc

Now that 1,200 employees have access to data via Tableau, I wanted to know: what's next? And what advice would they have for other organizations struggling to move beyond anecdotal on the one hand, and a proliferation of disconnected data sources on the other?

One step in the forward plans: more ad-hoc analysis, pulled from local data sources as well as BPL's data warehouse. Plunkett:

We also use Tableau as a tool for analysis when we're looking at a particular problem, or we're trying to answer questions that our stakeholders might have, or we are launching a new project and we want to understand who this might impact. We will also take the data that we have and do some ad-hoc, one‑off reporting that might answer the question of a local reporter or a government official. It might also help us understand what patrons are in a particular demographic, and what their activity is in the library as we're thinking of launching a new program. This level of analysis would have been much more challenging and time consuming in the past.

Plunkett believes that this ad-hoc aspect will spread through BPL. In turn, this fosters the culture change successful analytics projects usually have:

As more and more departments see what we can do with this, they want to be able to do it for themselves. The more we do about putting data out there, the more we can connect people who can share ideas. I think this is about a shift in how we want our culture to be here at BPL.

As for advice to her peers, Plunkett returns to the challenge of integrating and understanding the data. Like most large organizations, BPL has a bunch of data sources they need to consider, especially when it comes to one-off reporting, where everything from spreadsheets to Microsoft Access comes into play:

Pulling it all together is really the hardest part. Getting that data usable and clean is the biggest hurdle to overcome. Understanding that the data is the hard part - that's key to setting expectations about how quickly this can move.

If you haven't brought all your data together in the past, that's going to be a challenge both from a technical sense, but also from an organizational sense of understanding who should see this data. Should this be public? How can we share this and still feel comfortable that we're not putting out there what we shouldn't? You must get past people's ideas that "This data is mine," and shift to, "This data belongs to the organization, and sharing it will be beneficial to all of us."

The wrap: culture change and data access are linked

Looking ahead, Brooklyn Public Library plans to expose some of this data to their patrons, both as a communication aid and a way to engage their constituents with fun and/or useful information. Sharing that content should increase awareness of BPL's services as well. BPL is also looking at a storytelling feature in Tableau 8.2, which would allow them to frame their data into a visual narrative that would be appealing to their constituents.

There are plenty of ways to measure the success of such a project, including increased consumption, event attendance, book checkouts, and so forth. But for Leddy and Plunkett, the most exciting measure is when a light bulb goes off, somewhere in the organization, about possibilities still untapped. Plunkett sums it up:

What we love the most is when folks out in the branches call us and say, "This is really great, but I'm trying to figure out this other problem," or, "Because I saw this particular graph that you made, it made me think of anoother question. Is there a way that you can show me that?" We get the greatest satisfaction from questions like that, because it means people are looking at what we've built, and they're thinking about it deeply. They know their business in a way that we might not, so they're helping to guide us into what the next iteration of all this will look like.

Image credit: Feature photo: Library © Igor Mojzes - Additional photo provided by Brooklyn Public Library.

Disclosure: diginomica has no financial ties to Tableau - this interview was arranged by Tableau PR; I had complete control of the interview and editorial content.

A grey colored placeholder image