Advia Credit Union

Credit unions face mounting pressure to grow wallet share against banks and fintech lenders that invest heavily in personalization technology. Advia Credit Union needed to match the right financial product offers to the right members at the right time — a capability that larger institutions build with dedicated data science teams. Advia lacked that internal capacity, and its compliance obligations under fair lending laws added a further constraint: any member targeting effort had to withstand regulatory scrutiny. Without a scalable, compliant approach to personalized outreach, Advia risked either leaving loan opportunities on the table or inadvertently creating patterns in its campaigns that could trigger fair lending violations.

Advia partnered with Faraday, an AI platform designed for consumer-facing businesses, to deploy machine learning and predictive analytics without requiring in-house modeling expertise. Faraday generated custom predictive datapoints — specifically, next-best-offer scores — that ranked each member's likelihood to respond to an auto loan offer. These scores were fed directly into Advia's existing email and direct mail workflows, enabling personalized campaigns without rebuilding marketing infrastructure. Faraday's platform includes built-in bias mitigation features that evaluate predictions for disparate impact, allowing Advia's team to run targeted outreach while satisfying fair lending compliance requirements. The engagement began as a focused pilot on auto loans, giving Advia a contained, measurable test case before considering expansion to other product lines.

Within 90 days of launching the AI-personalized auto loan campaign, Advia reached a 5.18% application rate — a 3x increase over prior performance — and generated $2.7 million in new loans. Key outcomes:

• Application rate: 5.18%, up significantly from prior baseline performance
• New loan volume: $2.7M within the 90-day campaign window
• Compliance: fair lending requirements met through Faraday's integrated bias mitigation

Equally important, the results were achieved through existing channels and without adding data science headcount, demonstrating that mid-market credit unions can close the personalization gap with larger financial institutions using purpose-built external tooling.

• Piloting AI personalization on a single high-value product (auto loans) gives credit unions a clean, measurable proof of concept before committing to broader rollout.
• Choosing a platform with compliance features built in — not bolted on — removes the fair lending barrier that blocks many smaller institutions from adopting member targeting at scale.
• Predictive outreach through existing email and direct mail channels can triple application rates without new infrastructure investment.
• Third-party AI platforms with pre-built consumer data models can close the data science capacity gap for mid-market credit unions that cannot build or staff models internally.