Performance Perspectives Blog

Results not typical (too bad there isn’t a GIPS-type standard for all advertising)

by | Jul 26, 2012

While exercising this morning, I caught an advertisement on the television for a supplement to help men guard against prostate cancer. I had the sound off, so don’t know what was said, but at one point noticed the following:

Results not typical.

Difficult to read? It says “Results not typical.” Imagine trying to read it from across the room! It was in rather small white print, against a light blue background.

Yes, it was there; but clearly (no pun intended) not meant to be read by everyone, save for those who were sitting quite close to the screen. And yet their advertisement had testimonials which were apparently “not typical.”

To present performance of outliers would, of course, be an example of non-typical results. And while a distribution has outliers on both the positive and negative ends, I doubt that we will see an advertisement touting the results of those outliers that are on the left side of the curve.

GIPS(R) (Global Investment Performance Standards), of course, guards against this by requiring the inclusion of all accounts that meet the criteria the strategy being marketed; not individually, but collectively, with a dispersion method (if there are six or more accounts present for the full period) to provide even greater insights.

Many men over 50 will no doubt pay close attention to advertisements offering aids to guard against prostate cancer, and many (most?) will overlook the fine print, but will willingly pay for a supplement whose results may, in reality, not do much at all. Likewise, a manager who sells their money management services by highlighting the performance on the far right of the distribution will also be bringing on clients who will not normally achieve such results.

What this of course also means is that any time information is shared with a prospect that comes from a subset of the entire composite (e.g., a representative account’s attribution results), sufficient details must be shown so that what it truly represents is understood.

Free Subscription!

The Journal of Performance Measurement

The Performance Measurement Resource.

Click to Subscribe