The late 19th century Philadelphia department store owner John Wannamaker once said, “I know half of my advertising budget is wasted. I just don’t know which half.”

Maybe you feel this way about your advertising budget. Maybe you feel this way about all the money you spend trying to get customers to knock on your door and buy your products.

Certainly, if you look around during a recession like the one the pundits tell us we’re pulling out of now, you’ll see proof that a lot of your peers feel this way. Search through past issues of your industry’s trade journals: Many have shrunk or disappeared because of evaporating ad revenues. Read the local papers: Dozens of Portland advertising and public relations agencies have slashed their staffs or closed their doors because their clients have cut spending as sales have slowed down.

What does that tell you? It tells me that many companies treat advertising as if it were a gamble. “When times are good I’ll put a few bucks on the table and spin the wheel, but when times are tight, I’ll hang on to my cash, thank you.”

Measurement: The key to scientific advertising
Don’t approach advertising as a gamble. If you do, you’ll likely end up with self-doubt, second-guessing and sleepless nights. Approach it as a scientist might approach some problem in the natural world. Approach it as a scientific advertiser.

A scientific advertiser knows that every ad, every direct mailer, every promotion is a hypothesis: “The appeal I make through this program will persuade qualified customers to contact our company, eventually resulting in sales that justify the program’s cost.”

To test the hypothesis, a scientific advertiser runs the program and measures the results.

Like their white-coated counterparts, scientific advertisers test out their hypotheses on small samples of the population. That is, they don’t spend all their advertising budgets on one ad or one magazine. They run two different ads in the same publication and then measure the response to see which pulls in more returns. Or they run the same ad in two publications and measure the response to see which offers a better audience for their message.
When the scientific advertiser hits on a winning formula, he or she increases the sample size by adding new magazines to the list, always measuring results to ensure that the principle learned in earlier tests continues to hold true.

Resistance is futile
Many advertisers resist measurement because they think it’s too hard. They give excuses like, “We give people too many ways to contact us,” or “Our sales people won’t participate” or “We don’t have the right software.”

Think of it this way: Resistance to measuring response to your promotional programs is just as absurd as resistance to measuring revenue and expenses to figure out whether your business is profitable.

Measurement gives you the confidence that your marketing efforts really do pay off, and it gives you the confidence to continue investing in marketing even when nervous competitors stop. (This can help you build market share in a down economy, and that is very satisfying.)

And measurement is easy.

Measuring tools: The balances and beakers of the scientific advertiser
O.K., I lied. Measurement isn’t easy. But it isn’t rocket science, either. It just takes dedication. To be successful, you’ve got to stay on top of it month after month. Here are three simple tools you can use to measure the response to your ads and other promotional programs:

• Coupons: By including a special code on the coupon, you can tell which ad, brochure, flier or mailer the response came from. Coupons may be old fashioned, but they work. In one recent campaign, I watched 127 responses come in through faxed or mailed coupons. Only 39 came in through the Web.
• Special Web addresses: You can use your Web site to help track response. Print a special Web address like www.mycompany.com/specialoffer01 on your promotional material and then track how many people visit that page. Better yet: Send respondents to that page and ask them to fill out a form to request a bid or a special report.
• Asking: Whenever your company receives a phone inquiry, ask where the inquirer heard about you. This is tough because it involves personal initiative, but it helps provide a complete picture of response to your programs.

When you add the responses from coupons, Web sites and phone inquiries, you get a good picture of the response-generating power of your program. You’re almost there.

Measure what really counts: sales
Measuring response tells you how many people your program persuaded to contact your company. That’s a good start, but it’s not the whole story. Do responses pay your office manager’s salary? No. Do responses help you pay for new computers? No.

Response can be deceiving. That direct mailer you sent out might generate more inquiries, but the ad pulls in more sales. Remember, sales pay the bills. Sales generate profit. If you’re going to measure ROI, you’ve got to measure sales.

However, if the sales numbers are low, don’t be too quick to toss a program. Maybe it’s attracting the wrong people: Maybe all the respondents are after a chance to win the digital camera you’re giving away and don’t give a damn about your automated test equipment. If that’s true, try a new approach. Then again, maybe it is attracting the right people. Maybe your sales people wouldn’t know a good lead if it burst out of the stomach of one of their colleagues and started dancing on his chest. If that’s true, look into sales training.

Expect failure
If you're a scientific advertiser, you'll encounter many small failures along the way to success. That’s because every time you test a new program, it’s like you’re trying to feel the public pulse using acid-proof laboratory gloves.

But remember, the size of those failures is limited—because they’re restricted to experiment-sized samples. In the long run, the scientific advertiser’s small losses from failed experiments are overwhelmed by the gains from the successful programs that are run for larger and larger audiences until they wear themselves out generating profits for the company.

So what’ll it be? The roulette wheel or scientific advertising?