Skip to content

Case study · IntelliFi™ Predictive Analytics

From broad outreach to predictive growth.

How Penn East Federal Credit Union used IntelliFi predictive intelligence to sharpen acquisition, activate a multichannel campaign, and drive stronger checking engagement.

A credit union employee showing a smiling young couple their new account on a tablet in a bright branch lounge
1 · Predict4.1x top-decile lift
2 · ActivateMultichannel campaign
3 · Measure5.7x Rewards adoption
Client
Penn East Federal Credit Union
Solution
IntelliFi™ predictive analytics
Channels
Direct mail, digital, outdoor
Focus
Rewards Checking acquisition

The growth challenge

Acquire the right members, not simply more names.

Penn East wanted to grow membership through a rewards checking relationship designed to encourage sustained engagement.

The challenge

Traditional acquisition can spend heavily to reach people with little likelihood of responding. Penn East needed a more disciplined way to identify prospects, control media waste, and connect the offer to behaviors associated with a primary financial institution relationship.

The objective

Use predictive intelligence to prioritize likely responders, translate insights into campaign strategy, and measure whether the program created meaningful checking relationships, not just one-time account openings.

A connected intelligence-to-action approach

  1. 01

    Predict

    Rank likely prospects

  2. 02

    Activate

    Shape audience and creative

  3. 03

    Measure

    Connect outreach to outcomes

  4. 04

    Learn

    Feed results forward

Each campaign's results feed the next prediction.

Why IntelliFi mattered

The value was not a standalone score. It was the connection of modeling, strategy, creative activation, and attribution in one learning loop.

Predictive foundation

A model strong enough to guide action.

IntelliFi built a product-centric acquisition model using Penn East's existing relationship signals and a permissioned identity graph, then evaluated how well the model separated likely outcomes from random selection.

Overall model performance

79/ 100
Rated Excellent

As rated in the source model report.

How much better the model finds the right people

Positive outcomes found in each group, for every 10 found by random selection. Modeled performance, not campaign conversion.

1xRandom selection
~2xSecond-ranked group
4.1xHighest-ranked group
Read this chartA 4.1x lift means the highest-ranked group represented positive outcomes more than four times as often as the overall sample. It does not mean a 4.1x campaign conversion rate.

Campaign activation

Prediction became a coordinated growth campaign.

Penn East paired a compelling Rewards Checking offer with an IntelliFi-prioritized prospect audience and coordinated direct mail, digital, and outdoor support.

Observed adoption lift among new members

Checking baseline index1x
Campaign checking index1.2x
Rewards adoption vs. baseline5.7x
+20%New members acquired during the campaign adopted checking at a rate 20% above baseline, while Rewards Checking adoption reached 5.7 times its baseline rate.

Indexed presentation protects Penn East's underlying account volumes while preserving the reported performance relationship.

The Penn East FCU Rewards Checking campaign: a billboard, a direct mail postcard, and an Instagram story, each reading Say hello to checking that rewards you, with a $200 offer

Relationship quality

Engagement beyond acquisition.

The campaign's effect extended into checking behavior, including product adoption among new members and movement among existing members.

27.6%New-member Rewards adoptionObserved campaign outcome.
27.8%Share of existing-member checking openings that were RewardsObserved campaign mix.
6.7xNew-member Rewards adoption vs. existing-member rateComparison reported in campaign summary.

New relationships

The campaign attracted new members while steering a meaningful share toward the featured rewards relationship.

Existing-member movement

Some existing members opened checking or upgraded from a basic account during the campaign period.

Higher-value signal

The results indicate that predictive acquisition can support relationship depth, not only response volume.

The takeaway

Better targeting is only the start.

Penn East's results show the advantage of linking predictive intelligence to campaign design, coordinated activation, and outcome measurement. That connection turns a model into a repeatable growth capability.

01

Optimize for relationships

Define success around product adoption and engagement, not response alone.

02

Use intelligence across the campaign

Let prediction inform audience selection, messaging, and channel decisions.

03

Close the learning loop

Attribute outcomes and use the evidence to refine the next initiative.

Source: Penn East FCU predictive model and campaign attribution reports. Results are specific to this initiative. Public, data-safe edition: sensitive member counts, balances, audience thresholds, feature details, and targeting logic have been intentionally omitted.

Turn your data into your next growth decision.

See how IntelliFi would score your membership and market, or start with a conversation about where you want to grow.