Predictive analytics for credit unions: Real life examples of AI-powered marketing

Pre-pandemic, credit union brand champions would have told you their priorities were:

• Attracting new members affordable and sustainably
• Driving higher levels of engagement with existing members in more thoughtful and resonant ways
• Delivering on the promise of more fluid money management knowledge and access through technology

Today, those imperatives are not only still accurate; they have each become more critical to enterprise success as the organization looks to marketing to fuel liquidity needs without breaking the earnings model. In other words, do more with the same (or less) resources. For most, the solution is marrying proven traditional communication practices with emerging technology to increase and improve reach without substantial added cost.

Travel back as far as George Orwell’s 1984 and you’ll find the ominous rise of artificial intelligence played out in almost every dystopian future. Practical proofs of AI, however, have been the mundane tasks of the workplace that might have required micro-iteration over long timespans. Those of us in the advertising profession may have been lulled into a false sense of security, believing that AI/ML could never adequately account for the “creative process”.

With the arrival of AI-powered natural language processing tools like ChatGPT, the proverbial genie is out of its bottle for crowdsourced AI creative product. However, unlike new well-meaning but roughshod creative content examples, AI-borne predictive audience selection and message journey logic are time-tested and well proven. While AI may not have the finesse and context of the traditional warm-bodied marketer (yet) for content development, it is a clever utility to deliver more timely, relevant information to a more precise audience with predictive, automated pivots as behaviors change.

Artificial intelligence and machine learning have proven to add meaningful accuracy and efficiency to the database marketer’s tool chest. Credit union marketing teams and the agencies who support them have been asked to adopt nimbler, logic-based multichannel member journeys. AI accelerates both the development of the necessary “if-then” decision flow and the ongoing evolution of those flows that leads to the coveted member “segmentation-of-one”.

This is a remarkable sea-change for under-resourced credit union marketers! AI/ML is doing for the logic underpinnings of a member communication journey what Web 2.0 did for brand marketing (social media, pay-per-click, programmatic, etc.). AI means reaching more members with greater frequency without having to lobby for additional staff and budget.

There are solutions that streamline matched audience selection by tapping 400+ attributes for over 270 million Americans and aligning those personas to what you already know about your existing ideal members, to identify more of them. Then, AI can increase campaign confidence by using historic data about success outcomes to optimize reach and deliverability. This focus on data synthesis and behavioral patterns is frankly nothing new! Marketers have relied on this approach for decades, but Machine Learning (a form of AI) allows marketers to both accelerate and automate this work, freeing their team’s bandwidth and resources for more and better market strategy development.

Credit unions across the U.S. seeking to accomplish unique goals are discovering how artificial intelligence can add precision, speed, and efficiency to the campaign process. One such cooperative used AI to support its member acquisition goals when statewide reach suddenly presented both new opportunities and challenges.

Case Study: Gulf Coast Educators FCU ($1.27B, Pasadena TX) recently expanded field-of-membership to the entire state of Texas and was seeking new membership growth beyond its core geographic market. Like many credit unions awarded a broader FOM, Gulf Coast Educators wanted to tap into the new opportunity without slow retail expansion or expensive traditional mass media. Member Acquisition Likelihood Prediction was used to find potential members in the new FOM. Mixed Media was utilized (i.e. OTT, Video Pre-Roll, Pandora) to target high propensity individuals based on behavior and likelihood to act. A direct mail, email and digital targeting strategy to boost brand visibility with known educators in Texas was created. Marketing efforts were personalized and customized to the AI-powered personas to help increase engagement. The following results demonstrate the power of this “TraDigital” approach:

• 6x year-over-year core member growth
• 8x month-over-month core member growth
• 9% of new member shares came from expanded FOM market
• Organizational net member annual growth target of 2,200 eclipsed by almost 4x: over 8,000 new memberships originated during campaign year!

This is one powerful example of using artificial intelligence to capitalize on the accuracy and efficiency promised by emerging technology. The life of a credit union marketer made both simpler and more impactful in harmony. It turns out a machine-driven future doesn’t appear quite as dim as Mr. Orwell may have imagined it to be.