Kourtis: Testing Often Lacking in BTB
"The excuse is ... there's not enough universe -- I only have 1,000 prospects, so what can I do?" said Kourtis, who is president of the Bellevue, WA, direct marketing agency. "Our opinion is that you can go out and divide the thousand into two segments of 500 each, and you will not get statistically valid results but you'll get directional learning, and that's better than nothing."
Kourtis was a speaker at the "'Lead' Me to the Sale -- Winning Strategies for Closing the Lead-to-Sales Gap" session yesterday at The B-to-B Marketing Conference here.
"As business-to-business direct marketers, we have a responsibility to test more," he said.
His firm's "power testing" methodology allows for testing as many variables as possible.
"Where most direct marketers will test 'A' against 'B' against 'C' maybe at most -- 'B' wins, they go back out and test 'B' against 'D' and 'E' -- what we do is we test maybe 50 to 100 variables," he said. "We ignore statistical validity for the short term, and we look for hot zones where response is 100, 200, 300 percent higher. That allows us to find control programs very quickly for our clients."
He also discussed the importance of testing the big ideas, which he characterized as including list, offer, copy and package. They do not involve what he called "whispers," which include testing one color vs. another or one image against another.
"It's not bad to test very small things or 'whispers,' you just don't want to do it the first time," he said. "You want to test the big ideas first, get a baseline and then go out and test this other stuff."