Payment data is your richest signal of member behavior. Here’s how credit unions can use AI with mobile wallets, BNPL, and crypto to build member-centric banking.
Why Payment Trends Are Now an AI Problem
A quiet shift is happening in your portfolio: debit volumes are flattening, mobile wallet transactions keep inching up, and Gen Z members are approving BNPL plans faster than you can say “credit risk.” Most credit unions see these as separate data points. The smart ones treat them as training data for AI.
Here’s the thing about payment preferences: they’re now the clearest real-time signal of member behavior you’ll ever get. And if you’re not feeding those signals into AI models, you’re flying blind while big banks and fintechs are running full telemetry.
Inspired by insights from Tom Pierce, CMO at PSCU, and tying into this AI for Credit Unions: Member-Centric Banking series, this post looks at what’s actually changing in payments—mobile wallets, BNPL, crypto, “financial aesthetics”—and how credit unions can use AI to turn those trends into better experiences, stronger loyalty, and more sustainable growth.
The New Payments Reality: Frictionless, Mobile, Visual
The core shift is simple: members expect payments to be instant, invisible, and intuitive. AI is how you keep up at scale.
What members are doing right now
Across PSCU and industry research, a few patterns keep showing up:
- Mobile wallet usage is climbing fastest among Gen Z and Millennials. Tap-to-pay and in-app payments are default, not novelty.
- Buy Now, Pay Later (BNPL) is normal behavior, especially for 18–34-year-olds who may not trust or fully understand traditional credit.
- Crypto is less about spending and more about identity. For most members, it signals tech affinity and risk tolerance, not everyday payments.
- Financial aesthetics matter. Members don’t just want functional tools; they want interfaces that feel modern, visual, and personalized.
The mistake I see a lot of credit unions make? Treating these as “channels to support” instead of “data streams to learn from.” AI flips that mindset.
Payment behavior is the richest, most honest member data you have. AI turns it into context, prediction, and personalization.
Why this matters for member-centric banking
If your strategy is truly member-centric, then:
- You respond to how members actually pay, not how your core was configured in 2009.
- You use AI to connect payment data with lending, financial wellness, fraud, and service.
- You stop guessing which features to launch and start testing, learning, and iterating with real behavioral insights.
The reality? You don’t need to reinvent everything at once. You need to pick a few high-impact moves and wire AI into them from day one.
Mobile Wallets: The Front Door to AI-Powered Experiences
If members live in their phones, that’s where your AI has to live too.
How AI makes mobile payments feel personal
Every mobile transaction is a mini data point: where, when, how much, what category, what device. Individually, they’re boring. At scale, they’re gold.
Here’s how credit unions can use AI with mobile wallet and card data:
-
Next-best-action recommendations
Use transaction histories to predict what’s actually helpful next:- Offer a gas cashback promo to members who constantly fill up at the same station.
- Suggest a travel card upgrade when airline and hotel spend crosses a threshold.
- Trigger a check-in message when spending patterns suddenly change.
-
Smart, contextual alerts
AI models can distinguish between normal use and anomalies much better than static rules:- “We noticed a subscription doubled this month—want to review recurring charges?”
- “This looks like your normal grocery store but in a new city—are you traveling?”
-
Real-time financial wellness nudges
Combine account balances, spending categories, and historical trends:- “You’re on track to overspend your ‘dining out’ budget by 23% this month.”
- “You’ve paid $84 in ATM fees this year. Here’s how to avoid the next one.”
Most of this can be powered by relatively straightforward machine learning and decisioning models tied into your digital banking or card app. It’s not sci-fi; it’s orchestration.
Practical starting point for a CU
If you’re wondering where to start:
- Pick one segment: for example, members 18–34 who use mobile wallets weekly.
- Define one goal: deeper engagement with your primary checking account.
- Launch one AI-powered feature: such as category-based spend insights or automated savings nudges triggered by paycheck deposits and spending trends.
- Measure one primary metric: lift in debit/credit spend or increase in app logins.
Overcomplicating AI is the fastest way to stall it. Start narrow, learn fast.
BNPL, Credit Risk, and AI: From Threat to Opportunity
BNPL has already changed member expectations. The question isn’t “Will it last?” The question is whether your credit union will participate intelligently—or just watch balances migrate elsewhere.
What BNPL behavior tells you about members
BNPL usage isn’t random. It reveals:
- Cash flow stress (small-ticket BNPL to bridge paychecks)
- Credit avoidance (members who prefer fixed, known payments over revolving lines)
- Purchase priorities (which categories are worth financing—travel, electronics, medical, etc.)
AI models can turn those patterns into:
- Early warning signals for members at risk of delinquency
- Identified cohorts for responsible credit line offers
- Insights into which BNPL-like structures members actually like
How AI-powered CUs can respond
Here’s a sensible playbook for credit unions:
-
Monitor BNPL footprints via transaction data.
Members are already paying BNPL providers from your accounts. Model:- Frequency of BNPL payments
- Ratio of BNPL to income
- Categories funded via BNPL
-
Use AI to segment by risk and opportunity.
For example:- Low-risk, frequent BNPL users → candidates for a CU-branded installment loan or card with installment features
- High-risk, multi-provider users → candidates for proactive financial counseling or hardship options
-
Design CU-owned alternatives.
Member-centric banking means you build products that feel as simple as BNPL but with fairer terms:- POS-style installment options on your card
- Small-dollar, short-term installment loans with transparent pricing
- Auto-structured payment plans that members can simulate in your app
-
Feed everything back into your lending AI.
BNPL behaviors are predictive. Add them as features in your credit models and collections strategies.
If you’re ignoring BNPL because it sits “outside” your traditional credit lines, you’re missing one of the best behavioral data sources available.
Crypto and “Financial Aesthetics”: Signals, Not Sideshows
A lot of boards still debate whether crypto is a fad. Meanwhile, your younger members are using it to test, learn, and signal who they are financially.
What crypto interest actually means for CUs
You don’t have to become an exchange to learn from crypto behavior. AI can help you interpret:
- Risk appetite: members who transact in volatile assets may be more open to higher-yield, higher-risk traditional products.
- Digital comfort: crypto-active members tend to adopt new features faster across the board.
- Education gaps: transaction patterns often expose misunderstandings about taxes, volatility, and diversification.
That’s a content and advice opportunity. AI-driven recommendation engines can:
- Serve targeted financial education about volatility and long-term investing.
- Suggest safer, CU-backed alternatives that still feel “modern,” like thematic savings goals or robo-advised portfolios (where available).
Financial aesthetics: design as data
Tom Pierce has talked about “financial aesthetics”—the idea that members judge financial tools not only on function but on how they look and feel.
AI fits here in two ways:
-
Personalized UX
Use behavioral data to adjust what the app highlights:- Visual savings trackers for goal-driven members
- Clean, minimalist views for “just the balance” users
- Deeper analytics for data nerds
-
Smart content sequencing
Recommendation models can decide:- Which card art or virtual card designs to promote
- Which dashboard layout to default to for different personas
- Which notifications to show or suppress based on engagement history
Design isn’t just branding. It’s a feedback loop. AI is what makes that loop adaptive instead of static.
Bringing It Together: An AI Roadmap for Member-Centric Payments
If you strip away the jargon, AI for credit unions in payments comes down to three verbs: observe, predict, act.
Here’s a practical roadmap that respects limited budgets and staff:
1. Observe: Clean, connect, and label your data
- Consolidate card, ACH, and mobile wallet data into a usable environment.
- Standardize merchant categories and transaction labels.
- Tag segments: age, digital adoption level, primary relationship depth.
Without this, every AI initiative becomes a custom project. With it, everything gets faster.
2. Predict: Start with a few focused models
Pick 2–3 use cases that sit at the intersection of member value and CU ROI:
- Churn prediction based on declining card usage and wallet displacement
- Fraud scoring that blends rules plus ML, especially for card-not-present and digital wallet transactions
- Propensity models for:
- Enrolling in card controls
- Activating mobile wallet
- Accepting a balance transfer or installment offer
Keep them transparent. Your team should be able to explain, in plain language, why the model is making a recommendation.
3. Act: Orchestrate experiences across channels
AI without execution is just an expensive report. Use decision engines to:
- Trigger outbound messages through your app, email, or contact center
- Adjust in-app banners and offers dynamically
- Provide frontline staff with “suggested next conversation” prompts during member calls
For example:
“Member has shifted 40% of spend to an external BNPL provider in the last 90 days, but maintains strong direct deposit and low utilization. Surface CU installment line offer with pre-filled terms in mobile app and prompt MSR to mention during next interaction.”
That’s member-centric banking: relevant, timely, and actually useful.
4. Govern: Stay transparent and member-first
As you scale AI in payments:
- Document how you use member data and communicate it plainly.
- Build review processes for bias, fairness, and accuracy.
- Give members control over personalization and data use settings.
The differentiator for credit unions isn’t raw AI horsepower. It’s trust plus intelligence.
Where Credit Unions Go From Here
Most credit unions don’t need more data; they need better use of the data they already have—especially payment data. Mobile wallets, BNPL, and even crypto aren’t just trends from a PSCU report; they’re signals from your members about how they want to bank.
Member-centric AI means you listen to those signals and respond in three ways:
- Smarter, more personalized digital payment experiences
- Fair, transparent credit options informed by real behavior
- Proactive financial wellness support that respects member context
The institutions that win the next decade won’t be the ones with the flashiest app. They’ll be the ones using AI quietly, consistently, and ethically to make every payment interaction feel like the credit union knows the member—and is on their side.
If your team is mapping out 2026 priorities, start with one question: How will we use AI to turn our payment data into better decisions for our members? The sooner you answer that, the more room you’ll have to experiment, learn, and lead.