This is a tool used BEFORE any live, direct mail testing to pre-identify the best test ideas, those most
likely to compete...
The results identified one test family with real potential to beat the control and another with a
chance to be competitive...
This is a tool used BEFORE any live, direct mail testing to pre-identify the best test ideas, those most
likely to compete...
DonorVoice used the pretest tool with a sample of house file donors (to evaluate renewal implications)
and a universe of d...
of 4

Pretest tool case studies

Two recent case studies showing how non profits can drastically lower cost, improve net revenue and get to rollout faster with a revolutionary new testing protocol, one that identifies likely "losers" and "winners" BEFORE the time and expense of live testing.
Published on: Mar 4, 2016
Published in: Business      Technology      
Source: www.slideshare.net


Transcripts - Pretest tool case studies

  • 1. This is a tool used BEFORE any live, direct mail testing to pre-identify the best test ideas, those most likely to compete with and beat the control. Non-profits greatly reduce cost by NOT mailing test packages likely to perform poorly and increase net revenue by increasing volume on likely winners. The pre-identification of likely winners and losers is done in two parts 1) First, surveying donors who are representative of those who will receive the actual mailing, showing them visuals of the direct mail package and measuring preference using a very specific and battle tested methodology. 2) Using the survey data to build a statistical model to assign a score to every single element that was evaluated. This methodology is well established in the commercial sector and used by large, consumer companies (e.g. Coca Cola, General Mills, Proctor & Gamble) to guide product development for many of the sodas, cereals and detergents on grocery store shelves. Building a successful direct mail package is conceptually identical to a winning tube of toothpaste with size, shape, message and color considerations to name but a few of the moving parts you need to consider. A midsize, non-profit in the disease and medical research sector, had an acquisition control package with a family (whose child had the disease) who no longer wanted to featured. The organization faced a serious marketing challenge, identifying not just a new family to feature but one that could compete with the current control. They had several different families open to the possibility and a wildcard option of a researcher in the field. The cost to live test all the possibilities plus the control, even with their relatively small panel sizes (10k per), was going to be $46,000 and they only had budget to test 2 concepts against the control. DonorVoice used the pretest tool with a sample of house file donors (to evaluate renewal implications) and a universe of donors representing their acquisition target. We evaluated all the test options and, because this survey testing environment can accommodate far more “moving parts” than could ever be afforded or logistically managed in live testing, we also included prospective taglines the organization was considering as additional test elements.
  • 2. The results identified one test family with real potential to beat the control and another with a chance to be competitive. Conversely, it identified the rest as likely “losers”, including the one the organization staff considered to be best. The organization, with only enough budget for two test panels, elected to believe in the voice of the donor as represented in the pretest findings even though their own personal preference and internal, conventional wisdom would have led them to test different families. And the results?  The Pretest tool identified Test A (one of the families) having a high likelihood to beat the control and it did.  The Pretest Tool identified Test B as being competitive with, but still behind the control – exactly what happened in the live test.  Savings from not producing & mailing all 6 test packages against the control $46,000  Anticipated annual revenue increase with rollout of better performing control $72,000  Peace of mind with having a control beating, new acquisition package that would NOT have been identified without the pretest tool Priceless Contact info: kschulman@thedonorvoice.com www.thedonorvoice.com “We are extremely pleased with the information gathered from the DonorVoice testing, which allowed us to solve a tricky marketing dilemma with a reduced level of risk”
  • 3. This is a tool used BEFORE any live, direct mail testing to pre-identify the best test ideas, those most likely to compete with and beat the control. Non-profits greatly reduce cost by NOT mailing test packages likely to perform poorly and increase net revenue by increasing volume on likely winners. The pre-identification of likely winners and losers is done in two parts 3) First, surveying donors who are representative of those who will receive the actual mailing, showing them visuals of the direct mail package and measuring preference using a very specific and battle tested methodology. 4) Using the survey data to build a statistical model to assign a score to every single element that was evaluated. This methodology is well established in the commercial sector and used by large, consumer companies (e.g. Coca Cola, General Mills, Proctor & Gamble) to guide product development for many of the sodas, cereals and detergents on grocery store shelves. Building a successful direct mail package is conceptually identical to a winning tube of toothpaste with size, shape, message and color considerations to name but a few of the moving parts you need to consider. A large, well established non-profit helping children with disabilities applied the pretest tool as a back-test to validate several test packages for their year-end calendar appeal for renewal and acquisition. The organization spends a lot of money testing packages that are expensive to produce and fulfill given the use of front and back-end premiums. The cost of production and fulfillment is magnified with every test package that does not outperform the control, since money was“left on the table.” A mail package containing this many elements means an exponential number of choices being made, whether they are all acknowledged or not, about what to test. The inefficiency of A/B testing almost assures a sub-optimal combination of all these choices. Validating the accuracy of this tool against actual mail results means this organization can save hundreds of thousands of dollars from mailing less AND mailing “better” by only choosing packages empirically identified, from among thousands of possibilities, as having a chance to beat the control or deliver the same results for less cost (production and fulfillment).
  • 4. DonorVoice used the pretest tool with a sample of house file donors (to evaluate renewal implications) and a universe of donors representing the client’s acquisition target. We evaluated all the test package elements including outer envelope color, letter and reply form changes, labels with different gender based themes, calendar inserts and back end premium offers. The pretest tool correctly identified winning and losing test packages with only one exception. The tool correctly determined,  Outer envelope color matters to response but the test colors would not beat the control  Incremental calendar changes were no better than the control calendar  Back end premium offers beat the control with no back-end offer and that the two – duffle and tote bag – were not equal performers.  Outer envelope personalization beats the control and is the 2nd most important element after back-end premiums. Applying these findings before doing the mailings means the “losing” tests could have been avoided saving production time and money along with the opportunity cost of lost revenue for those mailed poor performing tests instead of control. Furthermore, volume for the predicted, winning test ideas could have been increased to take advantage of the better performance.  Savings from not producing & mailing predicted, losing test packages $45,000  Increase in net revenue by doubling panel volume on predicted winners $20,400  Increased year one annual revenue from getting to rollout volumes faster $257,000  Peace of mind knowing that, based on initial project insights, client can use the tool to find premiums with lower cost and high response Priceless Contact info: kschulman@thedonorvoice.com www.thedonorvoice.com “The DonorVoice pretest tool is a dramatic improvement to the testing process – we can test more ideas, faster and cheaper than live testing and get to rollout faster by putting more volume on the likely winners.

Related Documents