Predictive Modeling

A Leading National Bank Increases New Account Activity By More Than 33%


Download This Case Study


The Problem

A leading bank, with retail branches throughout the country, had been running a customer acquisition program that targeted new movers relocating near bank branches with a cash incentive to open new bank accounts.

Although the program had generated reasonably good results historically, the bank was anticipating reductions in available marketing budgets and needed to lower program costs while maintaining or improving overall results.


Data Models Target Highly Responsive New Movers


The Solution

The bank worked with Speedeon Data to develop a new mover response model. The predictive model identified influential attributes of responders and non-responders, and ranked prospects into 10 groups of approximately equal size, i.e., deciles, based on their likelihood of response.

The bank targeted the top five deciles, containing 54% of the “most responsive” new movers. 89,000 prospects were offered $150 for opening a new checking or savings account with a minimum opening balance.


Favorable Topline and Bottom Line Results


The Results

The direct mail campaign generated the following results:

New customer acquisition rates increased 25%. 2,243 new movers or 2.52% of the target audience responded to the offer and opened at least one new account, compared to 2.01% of the control group.

New account activity increased 33%. 4,424 new accounts were opened by the 89,000 targeted new movers, representing a 5.0% new account open rate, compared to a 3.7% new account open rate for the control group.

Account balances improved 37%.Targeted new movers generated $9.2 million in new deposits, compared to $6.7 million for the control group of comparable size.

Other metrics improved. On average, targeted new movers:

  • Opened 1.97 accounts compared to 1.86 accounts for the control group.
  • Deposited $4,100 into new accounts compared to $3,752 for the control group.


Significant marketing costs were avoided. By not to mailing to the least responsive 46% of new movers contained in the bottom five deciles, the bank was able to avoid approximately $40,215 in direct mail and data costs.


Response Models Optimize Performance Continuously



By focusing on the most responsive new mover targets, the bank increased new customer acquisition rates, new account open rates, and associated account balances, while avoiding significant marketing costs.

The bank used the response model to address on-going changes to marketing budgets. The capabilities to fine-tune new mover audience selects by likelihood of response and accurately predict acquisition campaign results enabled the bank to optimize program performance on a continuous basis.