Combining this format with an endorsement from Bill Haslam—the Republican candidate for governor—increased the likelihood of a vote for Ragan by a full 8 percent. Additionally, we found the more expensive self-mailer format had no advantage over a cheap postcard. We did the test mailing, surveying, and analysis for the Ragan campaign in less than five weeks. The optimum mailers were then sent out three times starting the second week of October.

After our tests, we predicted that Ragan would win with 52.8 percent of the vote. He won with 54 percent. And Ragan got his results for less than $2 a vote. The Hackworth campaign and the local media were blindsided by the results.

A Fundraising Approach

Along with message testing, we’ve seen results in the corporate world using the MVT process to help increase corporate donations and charitable giving in higher education. There are clear applications here for political fundraising, too.

One example comes from a large telecommunications firm looking to increase donations to their in-house political action committee. The company wanted to ramp up its PAC efforts in just five months to engage an increasingly competitive landscape, so it asked its PR pros to work with its lobbyists, in a combination not used before, to set up a series of fundraisers. They asked for our help. We tested these variables on a group of employees invited for the purpose of supporting the PAC but clearly told that their giving decisions were not part of their performance review.   

We involved their PR pros and lobbyists in talking about everything from the location of the event and the style of the invitation, to which refreshments to serve and who the pitchperson should be. The group came up with an initial list of 103 ideas that we then whittled down to 19 factors that were easy, fast, and inexpensive to implement. The team ran events for a week, testing different combinations of variables on only a small percentage of the employees.

As in Ragan’s campaign, many of the findings were counterintuitive. Serving alcohol and suggesting a level of giving both had a negative impact. Having the company lobbyist give the pitch with a basic script, but one they could infuse with a little personality, was actually the most effective. Our efforts in refining their message and their fundraising process helped increase donations by 238 percent, according to their numbers.

We employed the same process in a similar effort for Lincoln Memorial University, which tested 30 ideas involving their mail, email, telephone, and face-to-face solicitations. The experimentation identified a slew of helpful changes in the content, format, and timing of their mail and emails.

In the real world, good ideas are incredibly hard to separate from bad ones, and the benefit of being able to focus only on the good ideas is tremendous. It’s no different in the campaign world. Our experience shows that no one—executives, political consultants, professors, or subject-matter experts—is able to reliably determine which ideas are the helpful ones. By testing dozens of ideas at once, however, and then determining the 25 percent that should be implemented, positive outcomes are highly likely.

It was Mark Twain who said, “The trouble with the world is not that people know too little, but that they know so many things that ain’t so.”

By testing improvement ideas, we can avoid the hurtful and inconsequential ones. And the results are based on science, not intuition.

Dr. Charles Holland is CEO and Founder of QualPro, Inc., a Knoxville-based consultancy. The firm has conducted more than 16,000 business improvement projects with more than 1,000 companies, including many of the Fortune 500.