Use Data to Build Member Engagement
By: Melanie D.G. Kaplan
John Dorman has worked at the Texas Medical Association for 29 years, and for most of that time, business decisions were made, more or less, from the gut.
"We used to just base them on intuition," says Dorman, TMA's chief operating officer. And even after the organization hired a $6,000-a-week consultant in 2003 to learn about making data-based business decisions, Dorman wasn't convinced it would lead to any long-term benefit.
"At the time, I thought that $6,000 was crazy money," Dorman says. But the consultant helped TMA recoup that expense in one fell swoop. Turns out 1,100 nonmember physicians were buying malpractice insurance that should have been available to members only. And association staff members didn't discover this until they looked closely at the data. They then sent letters to the doctors that explained they would lose their insurance if they didn't pay their membership dues. Voila! They had 1,100 new members.
Through a combination of financial investment, education, and technology, TMA has made business intelligence—the transformation of raw data into useful business information—part of its day-to-day culture. Drilling down and finding information about certain members once took days. Now staff can do it in minutes.
Associations are sitting on a trove of membership data, and executives are starting to invest in exploring what is known as "big data." But it requires a larger commitment than simply hiring an expert or purchasing software and pushing a button.
Big Data, Small Steps
"Data presents a tremendous opportunity to make better decisions," says Elizabeth Engel, CAE, CEO and chief strategist of Spark Consulting and coauthor of a white paper, titled "Getting to the 'Good Stuff': Evidence-Based Decision Making for Associations," on using data to inform decision-making. "That said, there are certain things you need to do to use your data efficiently. The data itself doesn't tell you anything. It's all about finding patterns, creating hypotheses, and testing them."
Venturing into the world of big data can be daunting, no matter how big or small your association. Experts say it's important to have the support of senior management and to think about this undertaking as a culture change, not a project with a start and finish point.
"I think associations need to start looking at all the sources of data they currently own," says Beth Zemach, senior project manager at Association Management Center (AMC), which has a program called Informed Decisions that helps its clients gather and analyze data. "Once you assess what information you already own, you can start thinking about how it helps you achieve your strategic objectives, and patterns will emerge."
Stale. Messy. Inconsistent. Dirty. Call it what you will, but a lot of data collected over the years is not clean. Herein lies one of the biggest challenges for associations, says Peter Houstle, CEO of Mariner Management and Marketing. "Messy data means you can't compare analytics," he says. "That makes the data management process more difficult."
What exactly is messy data? It's data that suggests the impossible (for instance, that a member is 150 years old, or that somebody completed her residency the year before she graduated from medical school); data that is formatted incorrectly or inconsistently (such as Maryland spelled out versus abbreviated MD); outdated addresses; and duplicate records.
Often, messy data results from a lack of rules: What's a current member, what's a lapsed member, what's a prospect, and what constitutes an individual versus a company membership?
If these rules aren't documented, staff members will do things differently. "You need standards, you need to reinforce them, and you need an audit mechanism," says Elizabeth Engel, CAE, CEO and chief strategist of Spark Consulting.
When you find that colleagues are making mistakes—and they will, because less frequent users might not know the rules as well as membership staff—you want an audit trail, Engel says, so you can see which users are having trouble with which rules.
Often, when association executives discover the extent of their messy data, they get discouraged. "Associations will just throw their hands up at that point, before they even get started," says Engel. But she cautions not to give up so easily. Taken in small doses, and starting with the biggest problems, cleaning your house of data shouldn't be so scary. "The biggest mistake associations make," she says, "is saying it's too hard."—M.D.G.K.
She helps her clients understand why it's important to start with what she calls "small data." For example, in many organizations the education staff has some data, the marketing staff has different data, and even more data has been gathered from purchases and surveys. But to compare the data and use it to make decisions, it needs to be organized, and it must be clean (see sidebar).
AMC is currently piloting a program with one of its medical clients to standardize and correct existing demographic data. Now, when an individual joins, registers, purchases, or certifies with the association, a pop-up requires that person to update his or her information. "This allows us to get an accurate picture of the association members, and that's key in terms of growth and reaching more members and customers," Zemach says.
Rebecca Achurch, director of business solutions for Old Town IT, an association technology provider, says associations are looking at big data as today's "bright, shiny object" and wondering, what do we do?
"Don't get freaked out by it," she says. "I think most associations need to take a step back and think, 'OK, what makes the most sense for our organization?' " Achurch, who spent most of the last decade working for the American Chemical Society as AMS program manager, warns associations not to get caught in the "all or nothing" mindset. Rather, think about better data analysis and use as a building-block exercise and an initiative that becomes part of the organization.
It is possible—and prudent—she says, to start with baby steps. Her mantra: "Start small, start specific, and start with a robust set of data." For example, perhaps you want to increase attendance at a meeting or beef up volunteerism.
She advises associations to use a five-step approach to a data project: determine the scope, collect, clean, analyze, and communicate.
"If you're trying to increase retention, change your marketing campaign for a group that you think has lower retention," she says. "So you take a test, have a constant, and see if you can move the needle."
If you determine, for instance, that member retention lags after three years, but it's high for members who have been around for more than a decade, reaching out to folks between that third and 10th year is a no-brainer. "To me," Achurch says, "that's Marketing 101. You are using data to make strategic marketing decisions."
TMA's Dorman says getting started with big data doesn't require a big investment. Sometimes it's just a matter of understanding the capabilities of your existing software.
"If you can identify the data elements you're interested in and put those in a spreadsheet, then you can put it into an Excel PivotChart, slice and dice, and look at the data in a way you'd never consider if you were just looking at rows and rows of names," Dorman says.
Visualization is key in analyzing data—just think how much easier it is to comprehend a map of red and blue states rather than to see those states broken down in a list. Tools such as Tableau or QlikView help manipulate data and create visuals that show trends, retention rates, and growth areas.
"The technology out there is so much more sophisticated now," says Peter Houstle, CEO of Mariner Management and Marketing and Engel's white paper coauthor. "These new programs allow us to take lines of mind-numbing numbers and letters and see how data relates to other sets of data. And they allow anyone in the organization to ask questions, not just the IT guy."
The Pennsylvania Institute of Certified Public Accountants (PICPA) recently took another approach, working with Aptify, a provider of association management software, to create what Aptify calls Composite Engagement Scoring (CES). The system helps analyze member engagement and assigns a score to each person, based on things like leadership, meeting participation, and volunteering. PICPA began by targeting members with an engagement score of zero, otherwise known as "checkbook members."
"A lot of times we would guess about our strategy every year. We were good guessers," says Heidi Turley, PICPA's CFO and vice president of operations. "But you really can't argue with the data. So we decided we wanted to be smarter about how we were looking at data." She says the engagement score was a simple metric that allowed them to drill down and find out where these checkbook members (42 percent of PICPA's membership) live and work. Now, the goal is to engage them and change their scores.
"We're calling this the year of experimentation—trying programs outside of what we normally do," Turley says. For starters, targeting these members helps the CEO determine which member firms to visit.
Turley says it's easy to get overwhelmed. Even as a numbers person, she says she is sometimes amazed by the scope of the data. Her advice? "Keep it as simple as possible," she says. "Take little pieces at a time, don't go crazy with it, and have fun with the results. There's no right or wrong answer to this."
Melanie D.G. Kaplan is a Washington-based freelancer.
[This article was originally published in the Associations Now print edition, titled "Sift Skills" in the print edition.]
Kate Nevins , June 18, 2014
Thank you for this article. It is great to see that one can start incrementally with data mining. This is more do-able and less intimidating.