L’Oréal Paris, MaxFactor, and Their Strange Use of Statistics
95% of us would agree — but with what
95% of us would agree — but with what
In the small print of almost every ad that tries to convince you of a product’s effectiveness, you’ll see a favorable statistic. For example, 97% of 76 women agree.
Brands do this to show social proof. They use a celebrity endorsement and qualify a big statement like gives you noticeably firmer skin with positive statistical data.
L’Oréal Paris & MaxFactor aren’t the only offenders, but they are regular users of this marketing tactic.
Why Is This a Problem?
The issue with this kind of powerful statement is the strange sample sizes that these companies use. 95% of 342 agree.
In the example of this advert for L’Oréal’s Excellence hair dye, Eva Longoria shows her 100% grey coverage, and the claim is made of up to 100% grey coverage.
Ignoring the 5% of people this didn’t work for, the use of 342 is not an industry standard.
As a side note, it’s not even a round number. 95% of 342 is 324.9. I can’t help but express my skepticism.
You’ll see the sample sizes are strange in all cases on most adverts in the cosmetics niche. Check out the next Olay wrinkle cream ad, or this one for MaxFactor’s Lash Effect mascara below.
73% of 124 women agree. That’s 90.52 women. I’m not a doctor, but I’ve never met 0.52 of a woman, have you? Simple math suggests this data could have been taken from a larger sample size — and either favorable or unfavorable data has potentially been omitted.
How to Use Statistics in Your Best Interests — But Without Lying
Legally I have to say I can’t be sure this is how anyone’s data is being presented, but I can tell you that it’s possible for statistics to tell any story for any company. Irony dictates that 98% of you will agree with that, I’m sure.
A completely hypothetical example
Let’s pretend I’m selling shampoo, and my claim is it brings out your radiant, natural color and leaves hair noticeably softer.
I ask 100 people to try some shampoo. 63 of them, who got free shampoo, agree with my claim.
63% isn’t a persuasive statistic. Frankly, it’s terrible, so I reduce my sample size to the first 67 who responded and exclude the other 33 that I don’t like. Strangely enough, that 33% were mostly people who disagreed with me anyway. What are the chances?
Now 63 out of 67 people agree with my claim. That’s a whopping 94%.
This means that I can advertise that 94% of 67 people asked agreed with my claim, and it has much more selling power. The 6% that didn’t agree reinforce the believability of my data and the sales roll in.
After all, if I were going to cheat, I’d have given myself 100%, so 94% is much more credible.
Legally speaking, in this completely hypothetical situation, using a method I don’t think others use — I’m clean. Morally speaking, I’m simply a resourceful marketer.