Challenge:

A non-profit educational institution that has worked with Darwill for over 20 years required assistance to increase their donor base by targeting those who were most likely to contribute. The clients were previously spending a large part of their marketing budget on purchasing outside mailing lists with limited success. Darwill was tasked with building healthier donor mailing lists that would result in a rise of donations.

Solutions:

Darwill's Data Science team utilized the clients existing data to train and validate robust look-a-like and response models. Building these models would result in a more refined list of people who "look" and behave like current donors. Targeting people with similarities to current donors would achieve better response results compared to that of purchased lists. Anyone can be on a purchased list, even those who have no interest in donating, leading to wasted marketing on people who are not likely to respond. Through data models, we can filter these people out and only spend time marketing to those with a propensity to donate.

•  Train and validate robust look-a-like and response models with client data.
•  Identify potential donors. These people are more likely to donate based on similarities to existing donors.
•  Utilize refined lists to target better candidates with direct mail letter packs to optimize the client’s marketing spend.
•  Continuously measure campaign performance with match back reporting and iteratively improve.

Results:

Thanks to Darwill’s successful data models, we were able to reduce marketing spend while achieving a better outcome. The cost of this strategy was only 10% of what the client was previously paying to purchase outside lists.

•  Doubled the number of donors in the first year of implementation.
•  Reduced the time it took to produce a positive return on investment on the client’s acquisition campaigns  – from 3-4 years to 2-3 years.