Canadian Taxpayers Federation

How targeted digital advertising affected direct mail results

Experiment ID: #8421

Canadian Taxpayers Federation

A not-for-profit citizen's advocacy group dedicated to lower taxes, less waste and accountable government.

Experiment Summary

Timeframe: 10/26/2017 - 11/25/2017

The Canadian Taxpayers Federation wanted to test whether showing Facebook video ads to house file direct mail recipients, before and during the response period caused a lift in any key metric: conversion, donation size or total revenue. This test was based on the successful test executed by Hillsdale College.

The Taxpayers Federation created a control audience that was excluded from seeing any video ads, and a treatment audience that would see a rotation of three video ads (but no donation-centric or petition ads) for a one week period before the mail piece hit mailboxes and for three weeks after, the prime time in which a donor would respond. They chose videos that had been popular in the past, but were still relevant. They optimized the ads for reach—aiming to show ads to as large a percentage of the target group as possible. The goal of this test was to lift direct mail revenue—not to add an additional segment of online revenue or capture data through petitions. 

Ads were turned off once each video had an average frequency (number of times shown to each person reached) greater than 4.1. However, some videos were displayed as many as 4.5 times.

They spent just shy of $700 (CDN) to show ads to the treatment audience and waited to see the results come in.

Research Question

Will showing ads to a targeted group of direct mail recipients result in increased revenue from the treatment group?


 Treatment NameRevenue per VisitorRelative DifferenceConfidenceAverage Gift
C: Direct Mail Only $6.32$98.15
T1: Direct Mail plus Ads $7.9325.4% 92.6%$117.55

This experiment was validated using 3rd party testing tools. Based upon those calculations, a significant level of confidence was not met so these experiment results are not valid.

Key Learnings

It appears that with a 25.4% increase in dollars per visitor, at a 92.6% confidence level, that this was a success and should be expanded and replicated in the future. Overall, a $690 investment in Facebook ads, resulted in a nearly $10,000 increase in overall revenue.

Experiment Documented by Jeff Giddens
Jeff Giddens is President of NextAfter.

Question about experiment #8421

If you have any questions about this experiment or would like additional details not discussed above, please feel free to contact them directly.