Goldilocks Data: Benefiting from the data that’s “just right” might be easier—and harder—than you think.
Tom Shapiro – Cultural Strategy Partners
American Alliance of Museums “Alliance Labs” March 7, 2018
There’s big data, there’s your own organization’s internal data (aka hyper-local data), and, increasingly, there’s Goldilocks data—the “just right” blend of timely information, relevant context, and affordability. All three types can improve resource utilization and operational outcomes. Acquiring the data that serves you best, however, is challenging. What follows is a glimpse at the kind of data you might want and the hurdles to acquiring it.
Big Data and Hyper-Local Data
Let’s first mention and then quickly move on from big data. Big data[ii], made famous by the NSA, Amazon, and Facebook among others, requires massive investment to identify trends and build predictive models. In the museum world, big data influences everything from HR policy to exhibit design (e.g., from interpreting demographic trends to responding to the ubiquity of smart phones.) Culture-specific big data resources include the NEA’s Survey of Public Participation in the Arts, DataArts’ Cultural Data Profile (nee Pew Charitable Trust’s Cultural Data Project), Culture Track, and the National Museum Salary Survey.
This sort of third-party-generated big data is directionally valuable, but hard for museums to use in targeted, tactical ways. Knowing the average admission price nationally or even regionally, for instance, doesn’t help you handicap the financial and attendance implications of increasing your ticket price. While often these external databases can be sliced and diced to better fit your museum’s profile, results still tend to be more generic than you might prefer for specific applications.
On the other hand, hyper-local data—data collected by and from your museum—is exactly tailored to you. While necessary and useful, the perspective this data provides also has limitations. From attendance and fundraising to programming, collections, operating expenses and so forth, hyper-local data helps you track inputs and outputs from period to period and year to year. This helps you see your trend line, though only in comparison to yourself and without external validation or reference.
Finding the middle ground between these meta- and micro-datasets brings us to Goldilocks data. This most elusive, and perhaps most powerful data—the “just right” data—helps shine a light on what your museum is doing in the context of what other relevant peer museums are doing.
Multiple benefits emanate from this perspective. For example, you gain a reliable quantitative frame of reference that is right-sized to your museum’s scale and concerns. Also, in gathering this data you forge a group of museum peers whom you and your staff can call upon as needed. These twinned quantitative and qualitative resources can prove invaluable.
Perhaps the best way to understand this approach is through a real world example. The Northeast Small College Art Museum Association (NESCAMA) is a forum for museum directors from about 20 campus-based museums in New England.[iii] Sharon Corwin, director of the Colby College Art Museum, explained to me that she and her fellow directors leverage each other to share ideas, information, and questions on an ad hoc basis. If one member was considering off-site art storage, for instance, they could query the others on their experiences. A major drawback to this process, Ms. Corwin went on to explain, was that while the information worked for the moment, there was no mechanism to aggregate and archive the data in a way that could provide real benchmarks and trend lines. When someone else had the same question a year or two later, the query was repeated.
Three years ago NESCAMA engaged me to help develop and administer a more durable and reliable solution. Collectively we crafted a 100-question survey covering eight key topic areas: governance; budgeting; audience and attendance; staffing and human resources; physical space; collections, exhibitions, and programming; coursework and academic integration; and strategy. Our approach was to be both sufficiently broad to shine a light on a multitude of concerns and sufficiently brief to be practical.
The results now empower each museum to identify the most comparable museums for any given topic. Ms. Corwin reports back that, “the data-driven insights are credible and reliable because they come from respected peers. The information is useful both in addressing specific questions and expanding our view of alternative perspectives.”
Referring back to the previous example, another participating director told me that they recently shared with their college administration that their museum “ranked third in gallery space, but tenth in all other space. This supported our sense of not having enough back-of-the-house facilities.” They also know which specific museum peers to call on for more insights into this topic.
The data also helped the Williams College Museum of Art reconsider its governance structure. Discovering that many NESCAMA museums’ acquisition committees had both on- and off-campus representation led Williams to enlist its formidable network of art-savvy alumni to lend advice on important acquisition opportunities.
The data not only aids decision-making, it also helps non-museum experts—be they board members, funders, government officials, campus administrators, partnering organizations and others—more fully appreciate their museum in a relevant context. As Anne and Frank Goodyear, co-directors of the Bowdoin College Museum of Art, told me, “We engage with a wide range of supporters, including the college administration, faculty, students, parents, private donors, and foundations. Inviting these partners to study our peer data has become an integral and effective way to demonstrate how our Museum achieves its mission.”
Getting your Goldilocks data
As part of a recent institutional strategic planning project for the Speed Art Museum in Louisville, we conducted a peer survey of seven museums of reasonably similar profile.[iv] The results enhanced the Speed’s understanding of various approaches to organizational structure, budgets, audience engagement and other factors. In addition, the staff’s participation in the research developed new peer relationships that continue to generate exchanges of ideas and information. As this peer group persists, both the data and this sympathetic professional community will continue to strengthen over time.
Enlisting your peer set for a Goldilocks data program—contacting them, gaining commitment, forging mutual objectives—is just the first step. Key subsequent steps (and challenges) include developing your group’s specific survey content; securing timely and complete survey responses; scrubbing the data for quality; and, crucially, formatting the report for ease of use. For this you’ll need survey development, research, facilitation and other skills. Needless to say, this requires investment in active management and the fostering of communal responsibility among your fellow museums now and on an ongoing basis. Fortunately, the positive ROI for each user is almost immediate and increases from year to year. As a leader of the Williams College Museum of Art told me, senior staff has access to—and uses—the data on a regular basis to inform a wide range of conversations and decisions.
No doubt you can name 10 to 15 museums that share your museum’s profile and gestalt. You likely already look to their programming or have built ad hoc networks with many of them. Mining your group’s Goldilocks data could be another strategic way to establish more meaningful and productive relationships. This approach yields insights that improve decision making, strengthen funding requests, and help align staff and board around solid, relevant, actionable information. Getting data “just right” is hard, but it’s not a fairytale—and it can be a real plus in managing your museum.
ABOUT THE AUTHOR
Tom Shapiro is a museum strategic planning consultant. His firm Cultural Strategy Partners focuses on municipal and private museums, with an expertise in campus-based academic museums. He works with general, art, children’s, and culturally specific museums and foundations. TShapiro@CulturalStrategyPartners.com
[i] To simplify reading this article and at the risk of grammatical error, I will use data as a singular rather than plural noun.
[ii] Admittedly, I am conflating true Big Data (massive databases of user-specific information) with large-scale data surveys (anonymous results from big, long-term tracking efforts.) Virtually all Big Data efforts are in the private sector, such as FICO credit scoring, Money Ball-style athlete trading, and Google search algorithms. Health and government sectors tend to use more anonymous data to gain insights into issues ranging from disease tracking, K-12 graduation rates, and global weather trends.
[iii] NESCAMA participation is voluntary and has no dues or formal governance structure. Instead, it is a self-identified group of colleagues that share enough in common to warrant their uniting on certain efforts. The NESCAMA museums participating in the current data project are at: Amherst College, Bates College, Bowdoin College, Brandeis University, Colby College, Colgate University, Dartmouth College, Hamilton College, Middlebury College, Mt. Holyoke College, Skidmore College, Smith College, Wellesley College, Williams College.
[iv] Along with other factors drawing these museums together are that each is the primary general art museum in their region, have annual budgets ranging around $8 million and have surrounding populations averaging around 1.3 million people.