Subscribe
Share
Share
Embed
BlastPoint Deep Dive Podcast explores the intersection of data, AI, and customer insights. Each episode dives into how AI-driven solutions help businesses understand and engage their customers, improve satisfaction, and reduce churn. Join us for expert insights, real-world examples, and practical tips on leveraging AI to transform the customer experience.
Whether you're a business leader or a data enthusiast, this podcast will give you the tools to stay ahead in today’s competitive landscape.
If you want to learn more about what we do and offer, contact us today to schedule a demo at https://blastpoint.com/about/contact/.
Credit_Union_Survival_Guide__Hitting_ 65_Million_in_Deposits_wi
===
[00:00:00] Welcome to the Deep Dive. This session is, uh, a bit unique. It's really a collaboration between human strategy and well AI driven efficiency. We're exploring the huge power of data solutions specifically for the credit union space. Our mission today is to distill the absolute sharpest insights from a live webinar we hosted back in mid-September, was called From Loan to Loyalty.
Data-driven strategies for engaging every member. And believe me, that session was just packed with actionable intelligence.
Tom: Oh, it absolutely was a real powerhouse of practical knowledge. We're pulling insights directly from that discussion. It featured Tor Bernstein, who's the CTO and co-founder of BlastPoint alongside three, uh, really phenomenal practitioners in the field.
We had Josh Wilson, senior VP of marketing at Whitefish Credit Union Royce Nam, senior VP and Chief Marketing Officer for City Credit Union, and Steven Reed, who's the SVP of Marketing and Brand Development over at. P one FCU. So real hands on experience there.
Exactly. So settle in, we're gonna quickly give you [00:01:00] the strategic roadmap these leaders are actually using to turn their data into, well, deeper member engagement.
Hmm. Okay. Let's start with the, um, the market reality check that really forced this whole conversation. The sources we're drawing from highlight some key statistics shared at the Elon Credit Union Leadership Summit. And honestly, they paint a pretty challenging picture for cu.
Tom: Yeah, the challenge is sort of threefold.
You could say first there's this generational problem, get this less than 20% of Americans under 40 use a credit union at all. So you've got this aging member base that's, well, it's just not being replenished effectively.
And the second thing which really jumped out at me is that the satisfaction advantage CU has always had.
It's gone.
Tom: It's vanished. Exactly right. Yeah. Research from, uh. The financial research firm filing in shows that the historic member satisfaction edge that credit unions held over traditional banks, it's actually flipped. Banks are now scoring higher in satisfaction levels, and that just instantly neutralizes the credit union's, you know, longstanding [00:02:00] differentiator of having superior member service.
That's a big deal. Yeah. Wow.
Okay. And compounding that file also noted that 83% of credit union websites are now basically indistinguishable from bank websites. When I first read that, I mean, it felt like a really brutal indictment. You know, like they'd lost their identity online.
Tom: Right. But Josh Wilson, one of the panelists, had a really interesting, kind of clever take on it.
He argued that actually parody in digital services isn't necessarily a. Failure. It's, um, maybe proof of success in a way. It means credit unions have finally caught up on digital infrastructure, on capital deployment, on just having the digital conveniences members expect they're not seen as, you know, technological laggards anymore.
Okay, I see that point. But catching up only gets you to parody, right? If the websites look the same and if the banks are now offering good, maybe even better. Service. Mm-hmm. How do you differentiate then? It sounds like CU are kind of forced to chase this expensive digital infrastructure just to stay even.
[00:03:00] It's like running hard just to stand still.
Tom: That's exactly the strategic dilemma. Yep. And the sources make it really clear that the differentiator. It has to shift entirely to the experience. Digital services now heavily, heavily impact those satisfaction scores. We're talking net promoter scores, mps, which is, you know, the key metric for loyalty.
When your online banking goes down, or the ATM network glitches, your NPS just gets brutalized and to acquire those younger members. The under forties we mentioned the experience has to be seamless. The webinar really stressed this point. The online experience cannot look like, and I'm quoting here, A DMV form that we converted.
That's not gonna cut it. It needs to look and feel like signing up for TikTok. Mm-hmm. You know, immediate, frictionless, kind of engaging it.
Okay. That makes sense. Which brings us to the solution, I guess, because you can't just keep throwing money at the tech problem. The core insight here seems to be you can't win by just spending more.
You have to win by being smarter, smarter about who you serve and how you serve them. The whole discussion really boiled the strategic need [00:04:00] down to answering four crucial questions for any credit union. One, how do we win the right new members, not just any members? Two, how do we stop the churn, the loss of our profitable members, given that apparently only about 30% of members are actually profitable on average.
That's a startling. Number three, how do we grow share of wallet with the members? We do have. Finally, four. How do we boost efficiency and crucially reduce the cost to serve these members?
Tom: Right? And to tackle those four big macro challenges systematically. Steven Reed from P one FCU introduced what he calls the R-E-A-C-H framework.
It basically provides a blueprint, a structure for how to deploy your data assets against specific business outcomes. It connects the dots.
Okay. R-E-A-C-H. Let's break that acronym down for everyone listening.
Tom: Share 'em. So R stands for retain. Pretty straightforward. Preventing that costly attrition, keeping the members you want to keep.
E is for engage. That means getting members active with the right products and programs for them, not [00:05:00] just any product. The right product A is for acquire winning those new ideal members. We talked about the ones who fit your model. C is for Cultivate. This one's interesting. It's about supporting members through significant life events, like say, periods of financial trouble or maybe a major purchase.
And finally, H is for harness. That's all about optimizing your budget, your internal processes, making everything run smoother and more efficiently. And what's really fascinating here, I think, is that R-E-A-C-H forces credit unions to stop doing, you know, random acts of marketing. Instead, you align specific data tools to specific measurable business outcomes.
So like Cultivate isn't just about being nice, right? It means using predictive data to maybe anticipate major life events, a sudden change in income, maybe spotting a high risk financial dip before it becomes a crisis. Yeah, allowing to see you to proactively step in and offer specific supportive services when they're needed most.
That's powerful.
That makes perfect sense. It puts structure around it. And we saw the panelists kind of stake their claim across this [00:06:00] framework, didn't we? Steven Reed, who introduced it, he focused primarily on retain, engage, and cultivate specifically. He was targeting those, he. Indirect members, like auto loan holders to try and deepen the relationship, get more wallet share with sticky products like checking accounts.
Tom: Exactly. But the focus on efficiency was also really critical. Josh Wilson from Whitefish, he prioritized the acquire and harness buckets. He used this great analogy comparing in and Out Burger to McDonald's stressing the strategic necessity of focusing your marketing dollars only on those members or.
Potential members who are the absolute best fit for your specific products, rather than, you know, trying to be everything to everyone. Like McDonald's might be more like in and out focus,
right? Laser focus. And then Roy Nam from First City, he put almost all his focus, like 99% he said on acquisition. He called it the absolute lifeblood of growth.
So it really illustrates that strategic dilemma that I think every listener probably faces. Where do you focus your limited data resources? Do you double down on retaining and deepening your existing profitable [00:07:00] members, or do you go all in on highly efficient acquisition? It's a tough call. Okay. Before we get into how they actually did this, the specific playbooks we need to talk about the technology itself.
Ai, we saw that MIT research suggesting, what was it, 95% of AI projects fail to meet their goals. Yeah, that's huge. And part of that failure, I suspect is just misidentifying the tool for the job. Using the wrong tech.
Tom: Precisely. Yeah. The experts on the panel clarified this using something they called the AI nesting doll visualization.
It's a helpful way to think about it. So the largest outermost doll is analytics. That's the big umbrella, right? Collecting data, visualizing it, basic reporting. Underneath that, you have the term ai, artificial intelligence, which honestly has become such a buzzword. It's almost functionally meaningless.
Now, covers too much. The real core engine, the doll inside AI that's actually driving the R-E-A-C-H framework and delivering results for these CS is machine learning. Ml. ML is basically algorithms that learn and improve automatically when they're exposed to more data. And this is where credit unions are [00:08:00] finding real success.
'cause ML provides specific actionable answers, like it can predict exactly which member is, say 80% likely to respond positively to A-A-T-L-O-C offer. Or it can identify the top 10% of your members who are most likely to turn in the next six months. That's gold.
Okay. So ML is the workhorse. Got it. Where does deep learning fit into this nesting doll then?
I remember it was kind of flagged as potentially risky for cuus.
Tom: Yeah. Deep learning or neural networks. Hmm. They are incredibly powerful, no doubt, but they tend to operate as black boxes. Right. And that's the core problem for the financial industry. Especially regulated institutions like credit unions.
You need auditable explainable models for regulatory compliance, for fairness checks, you need to know why the model made a decision. Deep learning's often opaque. Nature makes it really difficult, sometimes impossible to explain or audit a financial decision based on its output. And that's often just an unacceptable level of risk.
Right. The explainability issue makes sense. And finally, the innermost all, I guess, is generative ai. The tools like chat, [00:09:00] GPT, that we all know and, uh, maybe use too much.
Tom: Exactly. That's primarily used for things like content creation, maybe rapid prototyping of ideas, building sophisticated chatbots for member service, which really falls under that harness bucket of the R-E-A-C-H framework, improving efficiency.
So understanding this hierarchy is pretty essential because it's really that ML layer, that machine learning capability, that's the necessary technology to actually execute the strategy defined by R-E-A-C-H.
Okay. That clarification is super helpful, and this feeds right into the, uh, the four step iterative data driven process that they said applies to every single R-E-A-C-H goal.
Right? I matter if it's retain, acquire, whatever. That's
Tom: the blueprint for execution. Yeah. Four steps. Step one, target. Use your ML tools to identify the right segments that could be your most at-risk members for retention efforts or maybe the highest propensity prospects for an acquisition campaign. Step two, understand.
Once you know who to target, you need to define the right message and the right [00:10:00] channel that will actually resonate with that specific segment. Personalization is key. Step three, activate, automate, launch the engagement of the campaign, automate it where possible. And step four, measure, iterate. And honestly, this final step is often the hardest, but it's critical.
You absolutely must adopt a scientific process. You test, you measure the results, and crucially, you have to be willing to accept failure. Sometimes use those results, good or bad, to refine the models, tweak the messaging, and improve over time. It's a continuous loop.
Yeah, the panelists really emphasized that point.
Since many credit unions might buy the same generic propensity models from vendors, it's that customized iterative testing, step four, that allows them to really tailor the data to their specific market, their specific members. That's the difference they argued between achieving success in, say, rural Montana versus success in hyper competitive Southern California.
You have to iterate. Okay, let's get to the proof then. So what does this all actually mean for the bottom line? Let's dive into the [00:11:00] playbooks and the specific results that the practitioners shared. This is where the rubber meets the road.
Tom: Absolutely. Let's start with case study one. Whitefish Credit Union and Josh Wilson, they were focusing intensely on efficiency, particularly during a market expansion,
right?
Their big challenge was expanding into a new de novo market, basically starting from scratch in a new area, competition was high, and they needed highly efficient acquisition spending. They just couldn't afford that old spray and pray marketing approach.
Tom: Exactly. So what they did was leverage ML models to target prospects on three different fronts simultaneously.
This is key. First, the likelihood of that prospect becoming a good member long term. Second, identifying their best product fit right from the start. And third, predicting their likelihood of responding to a specific channel, which in this case happened to be direct mail, doing all three at once, really refine their list.
The result was pretty transformative. Wasn't it
Tom: Huge despite reducing their prospect mailing list substantially because they cut out all the low propensity folks, their actual [00:12:00] conversion rates soared to 3.8%,
3.8% conversion on direct mail. That's that's massive compared
Tom: to typical rate multiple times the industry standard.
Yeah. It proved they achieved much greater efficiency and greater impact all while spending less total money on the campaign. That's smart spending.
Wow. Okay. Impressive. Next up,
Tom: next is case study two P one FCU, and Steven Reed. Their focus was different converting existing, but maybe low engagement, indirect relationships into truly profitable deposit relationships.
Ugh, the classic indirect auto loan challenge. Like many cu, they'd accumulated a large portfolio of these loans, which often don't produce much return beyond the initial loan itself. So the strategic goal was to convert those members into core account holders, right? Specifically getting them to open checking accounts
Tom: precisely.
So they used predictive modeling, again, powered by ml, focusing only on those high propensity indirect members. The ones the data showed were most likely to actually open a checking [00:13:00] account if asked, and they were careful about managing their communications. You know, avoiding message fatigue by not blasting everyone.
The outcome there, what did they achieve?
Tom: The numbers were significant. They saw a 3.5% increase in checking account penetration among that target group that translated directly to over 15,000 new checking accounts opened, and those accounts pulled in $65 million in new deposits.
$65 million in deposits is huge.
But what about the cost? Was it efficient?
Tom: This is the critical metric. Their calculated acquisition cost for each new checking account was roughly $75.
Okay. H how does that compare? Is $75 good?
Tom: It's incredible. Industry standards often run closer to $400, maybe even $500 per new checking account. When you factor in all the marketing costs, offers everything.
So by using ML to precisely target existing members who are already likely candidates, P one FCU achieved an acquisition cost that was about 80% lower than the industry average. That's the real power of applying R-E-A-C-H, specifically the retain and engage [00:14:00] parts using data, massive savings, huge deposit growth.
That's a game changer. 80% lower cost. Okay. Finally, case study three. First City, CU and Royce nm. His philosophy was a bit different, focused on AI as. Augmentation.
Tom: Yeah. His whole approach was centered on AI is augmentation, not skill expansion.
Mm-hmm.
Tom: Basically, if the technology could do some part of the work faster or better than a human, can you use it?
Let the tech handle it. Like if it can create a complex data model quicker or find patterns, you wouldn't spot lean into that.
And he showed some cutting edge examples, right?
Tom: He did. They use something called Egen ai. Which is kind of mind bending. It essentially means AI that can code or direct other AI systems.
They used it to create a real-time brand reputation dashboard, pretty advanced stuff. They also use standard ML to identify and really focus on their specific select employee group, their seg, which was LA County Public Employees, sheriffs, social workers, et cetera. And then they activated these highly targeted campaigns where even the voiceover in the ads was entirely AI generated.
[00:15:00] Wow. Okay. And his key takeaway, the philosophical point, it really shifted the focus of marketing, didn't it? He argued the goal isn't just selling the next product.
Tom: No, exactly. Uh, he argued the goal should be the next best conversation, not necessarily the next best product. It's a subtle but profound shift.
The next best conversation, explain that a bit more.
Tom: Yeah. He said marketing needs to act more like a conductor. Orchestrating a detailed member experience, an experience that ideally begins months before the member even thinks about consuming a specific product. He used the analogy of booking a Disney cruise line vacation.
Mm-hmm. How they manage that whole journey with incredible detail and anticipation long before you actually step on the ship. And achieving that kind of holistic predictive view is only really possible when your data tells you exactly what conversation is needed with which member, and precisely when.
That really broadens the scope of what data-driven marketing can be. It's not just ads, it's relationship management at scale. Got it. This has given us a fantastic sweep of the material we've gone from, [00:16:00] you know, those worrying satisfaction trends all the way to demystifying the tech layers like ML versus deep learning and seeing how that R-E-A-C-H framework really drives tangible results and efficiency and deposit growth through those case studies.
Tom: Yeah, we've covered quite a bit that competitive flip. Pinpointing the specific AI tools that are actually working for CU right now. Mainly machine learning and seeing real world results that dramatically slash acquisition costs and bring in deposits.
So to leave you the listener with a final, provocative thought to chew on.
Roy Snam suggested something really simple that kind of cuts through all this complexity we've discussed. He said, basically, don't wait for the perfect plan or the perfect tool or the perfect data. Just start banging rocks together. As he put it. Start playing with these AI tools. Now, even in small ways, the specifics of what you do first might not matter as much as the simple act of starting that iterative, targeted experimentation.
Just get started. Now this deep dive today really just served as a preview of what was an incredibly insightful [00:17:00] full session. If you want the full webinar, including all the detailed case studies we mentioned, plus an essential guide they shared on leveraging free external data, some ROI calculators, and even a personalized AI assessment report you can use, please find the YouTube link right there in this podcast description.
You can watch the full on-demand webinar anytime. Thank you so much for joining us for this deep dive. We really hope this has given you the knowledge and maybe the nudge you need to start turning your own data into deeper member engagement.