A credit union cut checking account acquisition costs by 30% by replacing a black box with a testing program it owns
A multi-billion-dollar member-owned financial institution broke free from agency opacity, built a hypothesis-driven direct mail program, and turned every campaign into a compounding source of learning.
30%
reduction in cost-per-acquisition vs. prior performance
33%
projected lift in response rate (0.12% → 0.16%)
4+
The challenge
The client, a member-owned regional credit union with several billion dollars in assets and a branch network across the mid-Atlantic, has spent decades building something a financial institution can’t outsource: member trust. It competes on relationships, and checking accounts are how those relationships start. Acquisition efficiency directly gates membership growth, which makes the checking program one of the most consequential lines in the marketing budget.
Direct mail was the primary channel for winning those checking relationships. But the program was run by an outside agency on a black-box model: the credit union handed over budget and data, and results came back. There was no readable test structure, no isolatable variables, and no way to tie outcomes to a specific creative, offer, or audience decision. Cost-per-acquisition couldn’t be attributed, so it couldn’t be managed. Every campaign was a one-off, and no learnings carried forward.
The stakes were about to get bigger. As the credit union’s addressable market expanded, every point of CPA waste would compound across a much larger prospect universe. Continuing to spend into an unmeasured channel meant continuing to lose ground—faster.
“We didn’t just need lower acquisition costs. We needed to actually see what was driving them, so we could keep improving campaign after campaign.”
– Client stakeholder, Regional credit union
The solution
Prowess replaced the black box with a transparent, hypothesis-driven testing program. Every mailing was restructured as a controlled experiment, with clean control and test cells sized for a reliable read, so the credit union could see exactly what moved response and cost, and compound those wins over time.
At the core is a hypothesis-driven testing framework that defines the variables worth testing (channel, creative, offer, audience) and structures each drop to isolate them. A measurement layer captures response and cost by cell and channel, rolling up to cost-per-acquisition so performance drivers are always visible. A predictive response model, built directly in Snowflake, scores and prioritizes prospects to focus spend on the households most likely to open a checking account. The same modeled audience is then activated across both direct mail and a matched email send through an ESP partner, with response capture tied cleanly back to source.
Snowflake serves as the single source of truth, centralizing prospect, campaign, and response data to feed both the model and the measurement. Instead of a program the agency understood and the client didn’t, the credit union now runs an acquisition engine it can see into, question, and own.
The result
Cost-per-acquisition on the checking account program dropped 30% compared with prior performance. Response rates are projected to lift 33%, from 0.12% to 0.16%, by concentrating spend on the model’s highest-propensity prospects. And the CPA improvement held up against control—confirmed, not assumed.
Just as important as the numbers: the marketing team now has full transparency into what drives response and cost. Decisions are faster and more confident because clean control-and-test reads let the team act on evidence rather than intuition. Budget risk is lower because every dollar is tied to a measurable, isolatable variable. And the framework is repeatable; learnings compound campaign over campaign, reducing dependence on any single outside agency.
What’s next
Prowess is now the credit union’s preferred partner for the direct mail acquisition program, with four campaigns delivered and more underway. Each cycle feeds the next: response data flows back into the predictive model, sharpening audience selection and pushing CPA and response efficiency further. Near-term, the program extends to more offers, audiences, and seasonal drops, widening the tested playbook across the acquisition calendar. Longer-term, the same tested, model-driven approach scales beyond checking into additional products and channels, giving the credit union a repeatable, transparent acquisition engine it owns.
About Prowess Consulting
Prowess Consulting helps organizations replace opaque marketing programs with transparent, data-driven ones. We build hypothesis-driven testing frameworks, predictive models, and measurement systems that turn every campaign into a source of learning, so results compound instead of resetting.
Learn how Prowess can drive measurable impact in your content and marketing operations
To protect client confidentiality, certain names, details, and identifying information have been modified or omitted. The results and outcomes presented are based on a real client engagement.
Prowess and the Prowess logo are trademarks of Prowess Consulting, LLC.
Copyright © 2026 Prowess Consulting, LLC. All rights reserved.
Other trademarks are the property of their respective owners.
Related Posts
A credit union cut checking account acquisition costs by 30% by replacing a black box with a testing program it owns
A multi-billion-dollar member-owned financial institution broke free from agency opacity, built a hypothesis-driven direct mail program, and turned every campaign into a compounding source of learning.
SCS automated a week-long quarterly workflow with AI that never leaves the building
An independent wealth-management firm gave its operations lead back a week of work every quarter using a local AI application that keeps every record on the lead’s own Windows® PC.
A California cemetery cleared a 15-year, $21.9M reconciliation backlog with auditable agentic AI
How a 110-year-old institution resolved deferred revenue across approximately 9,600 contracts in weeks, not months, with every figure traceable to its source.