Through a special arrangement, presented here for discussion is an excerpt from an article from the Joel Rubinson on Marketing Research blog.
John Wannamaker is famous for saying, "Half the money I spend on advertising is wasted; the trouble is I don't know which half."
Being a curious soul, I decided to calculate how much wasted advertising a marketer can afford and still have a successful campaign. It turns out that, for a campaign to pay out, less than one percent of impressions need to be "impact impressions," that is, directly lead to a purchase.
Let's work through the numbers with a prototypical brand:
- The brand: 10 percent of U.S. households buy it in a year and based on its price, repeat rate, purchase cycle and buying rate it annually generates 44 million unit sales and $110 million in dollar sales at retail.
- The action: It doubles its annual ad spend from $10 million to $20 million. At an average $5 CPM (cost-per-thousand), this means they are buying an extra two billion impressions across various media.
- The expectation: Generate a 20 percent increase in sales which translates into about 2 million more households buying the brand and generating 8.8 million in incremental unit sales.
- Impact impressions rate: It is a little less than one half of one percent which can be calculated as 8.8 million incremental units/two billion impressions. Alternatively, we can think of the incremental advertising as generating about two million more buyers (who then make a stream of purchases) which makes the impact rate even smaller.
Are the other 99 percent of impressions wasted? Not all are. They create consideration and familiarity from which a percent become new buyers; they reinforce beliefs and remind people of your brand as they are about to shop while competitors are still in consideration. But beyond that, I would advise against marketers thinking they can create such a rifle shot approach that could target the few percent of impressions that had an impact and save money by not buying the "wasted impressions." I think that there is a huge amount of randomness in consumer and shopper brains so the great majority of those billions of impressions is needed to allow a small percent of new buyers scattered about (in targeting and brand perception spaces) to accidentally find the brand and begin buying it.
I do think that targeting will improve results but can never eliminate what people wrongly call waste. Even digital, where we can use the most sophisticated targeting analytics such as retargeting and lookalike modeling of conversions, achieves lifts in response that might be multiples of what we would get otherwise, but are still in the single digits. This means target but don't over target — or, as Jack Wakshlag, chief research officer at Turner Broadcasting says, "Segmentation is like salt. A little bit is good, a lot can kill ya."