Revising the rules in measuring the effectiveness of an ad campaign – June 5, 2006
Growing up on the family farm, we had only two rules. The first: If you had a good crop, the price was bad. And second: If you had a bad crop, the price was good.
And I found that it wasn’t much better in the business world. The rule for advertising? I gradually learned that half of all advertising works. Trouble is, you never know which half! And that’s the rule.
So how can we make advertising more of a science, rather than an art? How can we measure the results that advertising dollars produce? And what metrics should be collected and used to guide future ad expenditures?
Digging for answers, I found that for 62 dealerships over the past two years, almost a third are wading through 4,000 people with their ad campaigns before they can muster one major unit sale. Yet others manage to sell that one unit after marketing to just 300 people.
The below chart shows the quantity of dealers that are at each level of market penetration. I have found that about 85 percent of dealer sales occur in their 60 most active zip codes. Analyzing those top zips, on the far left, we see that 18 dealers are selling one major unit (bike, sled, pump, generator, whatever) for each 4,000 people in their market area.
In the middle, we see that in one year, 23 dealers can sell one unit for every 1,300 persons available, and the next 13 dealers can sell one out of each 800 people in the surrounding area.
Then the stars begin to emerge. Three dealers require only 600 persons to find a buyer. Two are at 440, two more at 360, and finally, we find one dealer in West Virginia who sells one bike for every 300 of his neighbors!
Think of the costs. Think of the effort. Think of the time. One store wades through 4,000 and sells one unit, while another does it for every 300 people that drive down the street.
Now, there are a good many variables at work here. I looked at household income, average home value, percent white, percent Hispanic, percent black — you name it. I did find a small correlation between penetration and average income. (Believe it or not, as income went up, units sold went down!) But none of these factors was a high predictor of penetration or sales success.
The one number not available to me was the most important of all: average age. If your store is located in Sun City, Ariz., and surrounded by snowbird RVs with Wisconsin plates all winter long, you just might have to run through 4,000 people before you find a buyer. But, if you are in the hills of West Virginia, where a four-wheeler is how you go to the store and get home, well, one unit for every 300 folks is about right. Maybe even a little bit low.
What we have here is a new benchmark: major unit sales transactions per year per 1,000 in market population. Advertising techniques are not all the same, and some marketing minds are more creative and therefore more effective than others. Yes, they are all limited by the demographics and receptivity of the target market, (Selling refrigerators to Eskimos? Selling saddlebags to racers?) but we can now see just how much more effective one campaign manager is over another.
So here’s the new rule: one major unit sale per year for every 4,000 market population is expensive and bad. One sale for every 1,000 population is average, and one sale for every 300 market population is cheap and exceptionally good.
And sorry, but the answer is no. You can’t all move to West Virginia. And that’s the rule. psb
Hal Ethington has been associated with the powersports industry for more than 30 years. Ethington is a senior analyst at ADP Lightspeed. You can reach him at Hal_ethington@adp.com