Big Data: we hear about it all the time in the news and in our everyday lives. It is big, noisy, complex, and highly technical, but for thousands of companies all over the world, Big Data delivers business insights they could not have accessed in the past.

Big Data also may seem out of reach for most museums. However, the characteristics of Big Data are simple and easy to grasp, and these characteristics can be mapped to a similar and related concept called Small Data. Museums of any size can use Small Data to deliver a better experience to their visitors. By beginning with improving our understanding of Big Data, we can begin to harness the power of Small Data.

What is Big Data?

The concept of Big Data has been a relatively new development in the history of the human relationship with data, as it has been only a short while that we have had the computing power and tools to accomplish this work. Big Data is the collection, storage, and analysis of extremely large amounts of data. These large data sets are expensive to collect and maintain, and it takes a highly specialized skillset to conduct this work. But these efforts can reveal insights – connections between data sets and new action steps to solve a problem – that were not available to companies before the arrival of Big Data.

Let me illustrate the Big Data concept using a simplified analogy of bank ATMs. Let’s examine an imaginary large bank. I’ll call it NEMA Bank, just for fun. NEMA Bank owns thousands of ATMs all over the world. Every second of every day of every year, NEMA Bank customers use these ATMs to conduct many different types of banking transactions. Non-customers use these ATMs too. Every time someone uses an ATM, the ATM records information about the transaction: the customer name and card number, the type of transaction, the date and time, the ATM location, and how long the customer took to complete the transaction.

Imagine in your mind’s eye all of those banking transactions occurring at every second of every day. Imagine that NEMA Bank stores all of that information in databases. Imagine the constant flow of information coming at the bank from those thousands of ATMs used by thousands of people. The data in those databases: that’s Big Data.

The primary characteristics of Big Data are often referred to as the “Three Vs:” Volume, Velocity, and Variety[1]. By examining these characteristics, we can start to understand what Big Data can do, and how it connects with Small Data.

• Volume – Big Data sets are large volumes of data, so large that it takes immense computing power and sophisticated computer languages to control and analyze them. In the NEMA Bank example, NEMA Bank is storing an extremely large amount of data about ATM usage.
• Velocity – Big Data is created at extremely high rates of speed, and these rates change frequently. This characteristic describes how quickly new data are created, and how often those data streams change speed. The usage of the ATMs fluctuates during different periods of the day and year.
• Variety – Big Data sets can be comprised of many different types of data, both structured and unstructured. NEMA Bank collects a wide variety of data types from customer transactions.

Big Data = Capturing + Storing + Analyzing

In order to truly have Big Data, we have to turn the information into action steps we can use to make business decisions. Big Data is not solely the collection and storage of large data sets; it is also the analysis of these data to create new insights. If NEMA Bank simply collected and stored that information about their ATM usage, their data would not be considered Big Data. It would just be a big, expensive group of ones and zeros, sitting in a database gathering dust. NEMA Bank has to use the information it is storing for these data to truly become Big Data.

Analysis is where the power of Big Data is revealed. By analyzing those thousands of data points about how customers use its ATMs, NEMA Bank can make better business decisions for the bank and its customers.

A Better Day Out at the Zoo Because of Big Data

I was thrilled recently to catch an article published by Wired Magazine about a zoo using Big Data to improve their operations and visitor experience. The Point Defiance Zoo and Aquarium in Tacoma, WA, with about 700,000 visitors per year, struggled with predicting attendance and planning for staff levels on both sunny and rainy days. After employing a suite of Big Data analytics tools and databases, they layered NOAA weather data and historical attendance data together in one Big Data set to predict their attendance in advance, allowing them to set sufficient staffing levels for any weather. You can read more about this project in the Wired Magazine article.

Putting Small Data to Work in Your Museum

Even if your museum welcomes far fewer than 700,000 visitors per year, you can still use the Big Data approach to collecting and analyzing information about your organization, but on a smaller scale, by using Small Data.

Small Data projects allow us to apply the Big Data approach – analyzing a wide variety of data that are collected continuously over time and analyzed on a regular interval – to develop action steps that will help your museum to run more efficiently and to better meet the needs of your audiences.

Another way to grasp how Big and Small Data are connected is to think of Small Data as a chunk or piece of Big Data. A chunk of Small Data has the same characteristics as the Big Data set – a wide variety of types of information, and a changing speed and rate of data creation – but less volume.

Let’s consider a Small Data museum example. Pretend we have a small natural history museum with a goal of increasing membership in its local neighborhood. This museum has access to a large demographic data set about their region, and these data are collected regularly with a wide variety of information types included in the data set. The museum is interested in these data because it has a goal of engaging with its closest neighbors. By examining a smaller chunk of that database – a Small Data set with demographic information about particular zip codes – the museum can develop a picture of the audience close by and can develop action steps to reach solely those people. By taking a Small Data approach, this museum can focus on a smaller set of information – a cheaper, faster way of achieving their analytical and institutional goals.

This year I am helping a mid-size history museum in Massachusetts to undertake a Small Data project that will be ongoing throughout the year. By implementing a new survey tool and then analyzing the collected survey results, this museum seeks to understand how their target audience feels about the organization with the goal of delivering an even better experience to these visitors. This museum uses low-tech and low-cost tools: they do not yet have access to any specialized database or software, and we use paper-based surveys to collect the data. By implementing these surveys to collect a wide variety of visitor information continuously over time, this Small Data project will provide insights on year-over-year trends that the organization can use to measure changing visitor sentiment and to improve the visitor experience over time.

You might be wondering: how is this Small Data project at this history museum different from a visitor evaluation for an exhibit or program that my museum has undertaken already? Many evaluations of specific programs and exhibits also are examples of Small Data projects. This particular project is a visitor evaluation that follows the Small Data approach because it will be conducted throughout the year on an ongoing basis, it studies many different kinds of information, and the rate of collection will fluctuate throughout the year. The museum will compare over time how the results change as a way to understand changing visitor sentiments. But you can also run Small Data projects in your development, finance, and operations departments.

A final note about analysis

Data storage is relatively simple; deciding what to collect and how to analyze it to provide reliable results is more challenging. For a museum that has not yet undertaken any data projects, I recommend exploring resources to assist in planning and implementation before starting the project. Bringing in a consultant or hiring staff with these skillsets are two ways to start, and if that direction is not feasible, there are many other resources available for guidance within our professional community.

Big Data provides new power to achieve business goals. Museums can grab hold of this power through Small Data!

Emily Robertson is an Independent Museum Consultant based in Massachusetts who loves to talk about Big and Small Data. Reach out to her at emilyrobertsonmba@gmail.com.