3 Steps to Using Data and Predictive Models to Gain Members

We’ve reached a point in history where we’re surrounded by information. And what’s more, we now know how to harness this information to our advantage. Data is constantly driving further innovations in technology, healthcare, and education – just to name a few of many industries. Through gathering and analyzing data, we see remarkable patterns that can help us optimize performance and predict outcomes with a high degree of accuracy. Marketing, too, is more data-oriented than ever, as it aids in honing in on a target audience and in seeing what strategies are working and which ones aren’t.

‍For the professional association, data can be an enormous boost in helping to find potential members that will be a good fit for your organization, while simultaneously decreasing your marketing spend – giving you a better return on your marketing investment. Here we’ll cover some of the basics of data analysis and how it can be used to benefit your association and your members. Before we get into the nitty gritty of using data, let’s first get a bird’s-eye view of the process and its goals.

Fundamentals of predictive modeling

So most people have an understanding of what data is and that it can be useful. But, precisely how is it useful?

Well, when you’re trying to create better leads, you need to have an idea of what a good lead is. Undoubtedly you’ve had people join your association that were what you’d call “ideal” members: they paid their dues on time, they kept their membership for a number of years, they participated in events. Anyone would agree that these are the kinds of members you want in an association.

Now, if you look back on all these “ideal” members and start examining their various features and demographics, you may be able to see trends and similarities. From these trends, you can derive what’s called predictive models: data-informed models that give you an idea of the type of person that’s a good fit for your organization. You may have heard marketers call these customer personas – basically just more personalized forms of predictive models that are meant to portray your customer spectrum. Now, all of this may seem a little dense, so let’s break it down a bit further, starting at the first step: data collection.

Phase 1: Gathering data

Assuming you use some sort of CRM to manage communication with members, you have quite a bit of data on your members already. You know where they live, their ages, when they joined, how long they remain in your association, their career paths, the roles they play in their companies, and so forth. You're probably already using this data to track and monitor your members already. This data will all be useful for creating predictive models, as we’ll see in the next point. But the important thing to remember is, there is a direct correlation between how accurate your model is, and how much data you use to create the model.

‍It’s best if you can gather data over a period of several years (or, simply go back through your system and pull from past information) so that you get the widest span of information possible. Your past members are especially useful for this, as not only do they tell you important information about what type of person to target, but they also tell you about how long your average member stays with your organization – and may even show you a correlation between certain demographic traits and length of retention. Once you’ve tracked sufficient data, you’re ready to put together predictive models to help you hone in on your target audience.

Phase 2: Assembling models

Now comes the interesting part – creating models from your data. Most likely, you personally will not be creating predictive models, unless you’re some sort of math whiz. But let’s face it – we’re not in Good Will Hunting, and solving complex algorithms is a skill that’s well beyond reach for most of us.

‍So first, you’ll either use software or a data service to create your models for you. They’ll use a specific algorithm to map out certain data sets over a period of time, and in doing so will pick up on trends – however subtle they may seem to be, they’re still significant. For instance, your outcomes may show that you get a higher percentage of members from a certain region. There may be a number of reasons why this is the case (there may be fewer in-person networking opportunities in that region, there may be a school that puts out a large number of graduates in your industry, etc.).

But for now, all you really need to notice is the trend. Perhaps, not only do you see more members from within that region, but you notice that within that group, you have a large percentage of members that are within the 25-35 age bracket. Now, not only do you know where to look for new members, but you also know what age group to target. You can keep going ad infinitum, assessing various demographic traits to show you where and whom to target with your marketing messages. This is your predictive model: the result of past data trends that informs you of your best future outcomes.

Phase 3: Crafting personas

So now that we’ve seen how data can assist with finding new members, is there anything it can do to help with retention? The short answer is yes. It can actually, in essence, preemptively aid in retention. Let’s flesh out this concept a bit more.

‍When you created your predictive models, those models should have been formed using data on your past members’ retention rates. Where did you find members that renewed the most? What was their age group? What was their position in their company?

The characteristics of past members with the highest lifetime value (LTV) will help you craft customer personas. We mentioned these earlier. They’re a type of predictive model that focuses specifically on telling you what type of person will deliver the highest retention rate, and thus the most overall value for your association. Now, most likely you will need to form several customer personas, each representing a different demographic or segment within your membership population. But, this wide variety of personas gives you a diverse and yet very accurate picture of the types of people that are best suited for your organization. Not only does this aid with your retention rates, but it also helps your members as it tells you who where to find those people that will truly gain from their membership and reap all of its benefits over a long period of time.

‍Another thing your customer personas will tell you is what your past high-LTV members enjoyed most about your organization. What was it that got them to keep renewing year after year? Was it the events? The networking opportunities? Your industry insight? Whatever it was, this data tells you what the strongest aspects of your organization are, so you can give more of them to your members, and thus increase your retention rate.

Completing the picture

While hopefully your head isn’t swimming with these big concepts, we certainly do hope that this shows you the importance of data for membership – both in targeting and retention. Technology is shifting all the time, and wouldn’t you know it: tech is fueled by data. The more you adjust to utilizing and applying data, the better you set your association up for future adaptation. By evolving with the data trends of the times, your organization will not only thrive even under leaner times, but it will also be more successful at providing member benefits. And this, at the end of the day, is what it’s all about.