You are using an outdated browser. Please upgrade your browser to improve your experience.

In the latest episode of Libraries Transform Texas, host Michelia Mason, head librarian at Southwest Research Institute, welcomes Dana Whitmire, electronic resources coordinator at the UT San Antonio Health Science Center Library, to explore how data analysis and visualization can support better library decision-making.

Their conversation highlights practical ways libraries can use existing data to manage resources, communicate their value and better serve their communities, regardless of their size or technical expertise.

Listen Here

From evaluating collections to demonstrating community impact, libraries are discovering how data can help them make smarter decisions, improve services and tell a more compelling story about their value.

Libraries have always been in the business of connecting people with information. But increasingly, they’re also using information about their own operations to better understand their communities, evaluate resources and demonstrate their impact.

From collection usage and website analytics to service requests and patron feedback, libraries have access to an enormous amount of data. The challenge isn’t necessarily collecting more of it. It’s knowing what questions to ask and how to turn the answers into meaningful action.

Start with the Right Questions

Libraries collect data from a wide variety of sources. Collection statistics reveal which journals, databases and electronic resources patrons access. Service metrics track everything from interlibrary loan requests and reference questions to room reservations and instructional sessions.

Website analytics provide another perspective, showing how visitors navigate online resources, while national reporting and accreditation requirements generate additional information. Qualitative feedback helps explain the experiences behind those numbers.

But having access to data doesn’t automatically make it useful.

“Everything begins with a question,” Whitmire explained. “You can have buckets of data, and none of it will make sense if you’re not asking the right question.”

For libraries beginning to explore their data, those questions don’t have to be complicated.

Is a particular resource being used? Does it justify its subscription cost? Are library services reaching the intended audiences? Are there opportunities to improve access or eliminate unnecessary expenses?

Starting with a specific question helps librarians identify which information matters and what additional context they need to make informed decisions.

When Data Challenges Long-Held Assumptions

One of the most compelling examples Whitmire shared involved a major electronic resource package that her library had subscribed to for years.

Historically, the package was considered essential for a health sciences library. But over time, its subscription price continued to increase while important titles were removed.

Eventually, usage data revealed that only three or four titles within the larger package were receiving significant use.

Canceling the package wasn’t an obvious decision. Its longstanding reputation meant administrators might be reluctant to let it go.

Rather than relying on a single year’s statistics, Whitmire examined five years of usage and pricing trends. That historical perspective helped demonstrate that the package was no longer providing sufficient value.

The library ultimately canceled the larger subscription and retained only the few titles that patrons regularly used, saving thousands of dollars.

The example illustrates how data can help libraries challenge assumptions and make difficult decisions with greater confidence.

It also demonstrates why the story behind the numbers matters. A spreadsheet showing declining usage might not be persuasive on its own, but a clear explanation of changing costs, shifting demand and alternative options can provide a compelling case for action.

Make the Data Meaningful to Your Audience

Identifying an insight is only part of the process. Librarians also need to communicate what they’ve learned in ways that resonate with administrators, colleagues and other stakeholders.

For Whitmire, that means recognizing that not everyone wants to spend time interpreting spreadsheets.

“Not everybody loves a spreadsheet as much as I do,” she said.

When presenting information to decision-makers, simplicity and relevance are essential.

“Our administrators have 30 seconds. You have to quickly catch their attention,” Whitmire explained.

A useful visualization should answer a specific question and present information in a context the audience understands.

That might mean highlighting financial savings for budget-conscious administrators, showing changes in service demand for library leadership or demonstrating how resources support research and student success.

It also means avoiding unnecessary library jargon.

Rather than overwhelming stakeholders with detailed tables and technical terminology, librarians can use charts, dashboards and concise explanations to communicate what the data reveals and why it matters.

Dashboards That Improve Services, Not Just Reporting

Data visualization can do more than help librarians explain their work. It can also make library services easier to access.

Whitmire described a dashboard her library developed to help faculty, staff and students identify journals and publishing packages eligible for read-and-publish discounts.

Previously, that information was maintained in spreadsheets, and users often needed to contact a specific staff member to find out which publishing opportunities were available.

The process depended heavily on individual knowledge and wasn’t particularly efficient.

By developing a dashboard that displayed eligible titles, subject areas and available publishing discounts, the library made that information more accessible.

“It allowed the service to grow, more people had access to it,” Whitmire said.

The dashboard also evolved to include information about savings generated through the program.

This created an additional opportunity to demonstrate the library’s value, not simply by reporting how many people used a service, but by showing the financial benefits those users received.

It’s an example of how thoughtful data visualization can improve both the patron experience and the library’s ability to communicate its impact.

You Don’t Have to Be a Data Scientist

For librarians interested in doing more with data, the prospect of learning new software or building dashboards can feel intimidating.

Whitmire’s advice is to start with familiar tools.

“Start with Excel. Play with pivots, tables. They will save you time,” she recommended.

Excel offers a practical starting point for organizing information, identifying trends and creating basic visualizations without requiring specialized software or extensive technical training.

Whitmire also uses Power BI, which is available across her campus, to develop more sophisticated dashboards and reports.

Learning the platform took time. She relied on YouTube tutorials, ChatGPT, colleagues and informal opportunities to experiment with the software. It took approximately six months to a year before she felt comfortable working with the data and producing the visualizations she wanted.

Even now, she continues learning through a campus Power BI user group and regular collaboration with colleagues.

The lesson for libraries with limited resources is that meaningful data analysis doesn’t have to begin with an elaborate system.

Starting small, building confidence and developing sustainable processes can be more valuable than attempting to create a sophisticated dashboard before understanding the underlying information.

Numbers Don’t Tell the Whole Story

Although data can reveal important trends, Whitmire emphasized that numbers must be interpreted carefully.

Different vendors may report usage differently, and information supplied in a report may not always match what librarians can retrieve directly from a vendor’s dashboard.

Understanding where data comes from and how it is calculated is essential to making reliable decisions.

Privacy is another important consideration.

Whitmire explained that her library works with aggregate data, limits access to sensitive information and de-identifies information when necessary.

Just as important is understanding the context behind the numbers.

A resource with low usage might initially appear to be a candidate for cancellation. But if it supports a small, specialized academic program, its value may extend well beyond the number of downloads it generates.

“Context is crucial,” Whitmire said. “We have to look at the numbers, and then also think about how it’s being used.”

That distinction matters for every type of library.

High attendance doesn’t automatically mean a program achieved its goals. Low circulation doesn’t necessarily mean a collection lacks value. And some of the most meaningful library outcomes may be difficult to measure through traditional statistics.

Data should inform professional judgment, not replace it.

Moving from Activity to Impact

Looking ahead, Whitmire sees an important shift in how libraries will be expected to demonstrate their value.

“I think libraries will increasingly be asked to show impact and not just activity,” she said.

Counting visits, downloads and transactions will remain important, but administrators will increasingly want to understand what those activities accomplish.

For academic and health sciences libraries, that might mean connecting library services and resources to research productivity, student success and patient care.

For other libraries, the outcomes may look different, but the underlying question is the same: How is the library making a difference in the community it serves?

Whitmire’s library is already exploring ways to communicate that value more effectively, including developing website pages dedicated to collection investment, service impact and resource allocation.

These pages will incorporate visualizations that help users understand how the library invests its budget and supports its community.

She also anticipates that artificial intelligence will play a growing role in data analysis, particularly in helping librarians clean information and identify patterns. However, she cautioned that AI-generated results still require careful review and verification.

Collaboration will also become increasingly important as libraries share information and identify opportunities to improve services together.

Every Library Can Start Somewhere

For libraries interested in becoming more data-informed, the most important step may be recognizing that they already have valuable information at their fingertips.

A collection report, website analytics dashboard, spreadsheet of service requests or set of patron comments can provide a starting point.

The key is identifying a meaningful question, examining the available information and considering what the findings reveal about the people the library serves.

“Data is amazing. It’s not scary. It can feel scary, but it’s not,” Whitmire said.

Ultimately, data analysis isn’t about producing more reports or filling dashboards with impressive charts. It’s about using information to make better decisions, allocate resources thoughtfully and communicate the value libraries bring to their communities.

And sometimes, the most powerful insight begins with a simple question.

Listen to the full episode of Libraries Transform Texas to hear Michelia Mason and Dana Whitmire discuss practical approaches to library data analysis, visualization and demonstrating impact.

Libraries Transform Texas is a podcast produced by the Texas Library Association, highlighting the work and value of Texas librarians and libraries.