Today’s post comes from Bridgz statistical analyst Paul Edwards.
Here’s a reminder for all of us in the marketing mob. Every time I hear one of those insurance TV commercials touting that individuals can save $411 a year on average if they switch to brand X car insurance, I want to rip my hair out. Equally annoying are those ads that state that a person can save, on average, $795 a month if they switch to a “new and improved” drug prescription plan. Now maybe I am just a stats nerd, but for an individual to make a decision based solely on the average (in any scenario), seems absurd to me. Here’s why.
Let’s assume the insurance carrier in the TV ad provided us with the amounts of money people actually saved when they switched to their awesome car insurance (see table A). Notice that the average yearly savings for the nine individuals is $411. Great, let’s switch! Notice also that the standard deviation of the data (relax, standard deviation is an easy statistical concept) is $27.50. This amount gives an individual an idea of how spread-out the data is from the average. The larger the standard deviation the more widely the data varies. In scenario A, a standard deviation of $27.50 indicates that the yearly saving amounts for all nine participants were relatively close to the average of $411. Based on these figure (average and standard deviation), an individual is at least justified in looking into switching insurance carriers.
However, let’s assume the average yearly savings for the nine individuals looks a bit different (see table B). Notice that the average savings is still $411, but the standard deviation has increased drastically from scenario A –$27.50 to $539.05. This indicates that in scenario B the saving amounts are varying much more substantially from the average. An individual may still save $411, but their chances are lower than in the first scenario. An average yearly savings of $120 seems more probable in the second scenario, based on the figures provided and assuming a non-biased sample.
It’s a good practice for marketers to remember that average values tell an individual only half the story. Some companies (not just insurance) are betting that customers will change/switch/buy based on only this statistic. Using these tactics, the company might make a sale, but will it build customer loyalty once the person finds out how little they’ve saved?
Knowing how dispersed the data is (aka standard deviation) tells the other half of the story. Together, averages and corresponding standard deviations will help marketers make more accurate and credible claims.
1 Comment so far
Leave a comment