t seemed like a no-brainer: Tire racks in front of Firestone Complete Auto Care stores would help help up greater sales. it turns out, however, that those displays were representative of something else: The fact that no matter how good an idea seems, a gut reaction about its potential for
Road Testing Ideas
MVT helps Bridgestone/Firestone find programs with real traction
BY FIONA SOLTES
"I no longer trust my own opinions," says Dave Cochran, vice president of operations for QualPro. And he's far from alone. A QualPro strategy called Multivariable Testing (MVT) uses high-level statistics to show which ideas --- and combinations of ideas --- can actually make a difference in a company's bottom line.
More often than not, however, MVT demonstrates that the things that "should" work, don't. Not only can ex-pensive programs have minimal impact; some can even cause more damage than good.
"It doesn't matter who it is," Cochran says. "The greatest expert can't separate the good ideas from the bad. We're
success holds virtually no value at all.
learning that the world is a really com-plicated place."
Bill Perdew, COO of Bridgestone sub-sidiary BFRC, addresses the annual QualPro Leadership Symposium in October. The audience was all ears, especially since Perdew shared that the MV process had resulted in 8.1 percent growth across its more than 2,300 com-
pany-owned stores.
It also saved the company more than $600,000 that was to have been spent implementing a post-service "thank you" hangtag strategy many thought would be a sure-fire hit.
When you ask Bridgestone/Firestone customers what they're looking for, "trust and convenience are the two keys," Perdew told the group. "One thing that I'd like everybody to remember is we're not the place [people] want to go." When polled about having their car serviced, customers said "they'd rather do their taxes, renew their driver's license, to to traffic court or --- the one I really love --- they'd rather go to the dentist," he said.
"When they drive into our facilities, it is an expense. It's probably not budgeted. It's probably going to have a negative impact on their grocery budget, so it's not a pleasent experience. Our challenge in our industry is trying to make it as painless as possible."
It should come as no surprise, then, that perdew found himself "intrigued" by the possibilities that Knoxville, Tenn.-based QualPro offered. The idea was to take the guesswork out of the equation and replace it with good, hard facts.
In the fall of 2006, Perdew and his
56
STORES / DECEMBER 2007
team began holding brainstorm-ing sessions across the country. Everyone from tire technicians to vice presidents was included, all with the goal of gathering ideas that would be fast, practical and cost-free to implement.
The initial list of 829 proposals was eventually sifted down to 36. At that point, they included things like follow-up calls to customers,market-specific store hours, female clerks and part-time staffing on Sundays. Those variables were then mixed into unique "recipe" combinations for testing at 128 stores --- a handful of changes at each one. After implementing those recipes for six weeks, the mathematical analysis began.
MVT
is "a way that
companies can
mine the creativity of
their entire workforce,
and do so in a way
that's very efficient."
- Dave Cochran,
QualPro
As it turned out, BFRC was spending a lot of money on things that had virtually no impacts --- and the company isn't alone. QualPro has found that only 25 percent of ideas tested through MVT --- no matter how strong those concepts seemed at the outset --- make a positive difference. Another 53 percent make no difference at all; those that re-main actually have a negative impact.
"Probably the biggest surprise for me was that there's only a 2 percent predic-tive factor," Perdew says. "When you survey people and they give you ideas that you believe wil work, and then look at the results, there's only a 2 percent correlation. It surprised me how far from reality our perceptions are. And I don't think you can find that out any other way."
Lessons learned
BFRC, which was already in the process of re-imaging its stores, learned that window graphics have no effects, even though they'd been hotly contested within the company. They learned that female clerks are more likely to stay at the counter ready to wait on customers, rather than migrate back to the service bay to lend a hand. And they learned that shifting direct mail pieces to later in the week --- and adding new ZIP codes --- provided a huge return.
And the process itself brought an added bonus; a boost for the company's workers, who were happy to offer their thoughts.
"Our CEO has often said that MVT is the best morale-building methodology in the history of mankind," Cochran says. "It breeds a culture of empower-ment, and the rank-and-file get really excited when they find out people are coming to ask their ideas. What MV does is focus on the impact of the idea rather than its source, and that's really exciting. It's literally a way that compa-nies can mine the creativity of their entire workforce, and do so in a way that's very efficient."
Perfect candidates
Instead of simply bringing in a con-sultant to discuss what has worked else-where or running tests on one "good" idea at a time, MVT allows companies to find unique synergies that they might not otherwise have considered, he says.
"When we do two things together, we find that there may be a different impact than when we do them individually," Cochran says. "Think about taking two medicines, for example; it wouldn't be unusual to find that one might effect the efficacy of the other."
Because MVT is successful at win-nowing out the truly beneficial ideas, Cochran says, some companies use the process over and over. "People are simply amazed by the size and impact that can be accomplished."
But therein lie a challenge. Sometimes, company executives get so excited when they see the test results that they believe they can improve the picture even more by tweaking something here
or there. And those little
changes can have signifi-cant unintended conse-
quenses.
"A QualPro client coined the phrase, "Test what you'll im-
"But what's even more tempting is to see the results of something that works and say, 'We can make this even better when we roll it out.' There's just as great a chance that the change will destroy that finding as enhance it. Sure, you have ideas to make it better," he says, "but let's test those ideas."
For his part, Perdew says, "there are things today that I won't make a decision on without running an MVT. My advice is, don't question the process, and more than anything else, don't question the results. This allows you to feel confident about where you're putting your time and emphasis."
Fiona Soltes, who splits her time between "retail therapy" and freelance writing, lives near Nashville, Tenn.
STORES / DECEMBER 2007
57
25 years. Retail op-erations are ideal for MVT, Cochran says,
because the activities typ-ically identified can result in
QualPro has tested
more than 100,000
variables for more
than 1,000 compa-
nies over the last
comp sales of 5 to 15 percentage points.
plement, and implement what you've tested," Cochran says, "There is always that temptation to ... include things that can be tested but would be really hard to implement. We try to head those off.