Numbers and Testing Trump Politics

I've encountered many situations where decisions have been made and policies created by people who seem to be operating from a distance, or who clearly have allowed personal feelings to sway their thinking.

Often, the better course of action can be obvious to lower-grade employees or outside consultants who may be ignored, discouraged, or even intimidated from offering their suggestions.

In this scenario, clear results from user testing and reviewing web traffic information can be very helpful in allowing the objective presentation of facts in support of recommendations.

Several years ago, I reviewed the website of a large professional association. The site opened with a splash page (that is, a page whose only function was to present a link that said "Click here to enter"). This page was frankly hideous. However, it had been designed by the nephew of the Director of Marketing, and the doting uncle refused all requests to order it removed.

My audit of the traffic reports showed conclusively that 30 percent of visitors left the site immediately upon seeing the splash page.  Clearly this was a significant number (and interestingly, one that I found pretty consistently to be the bounce rate from splash pages). The Director of Marketing couldn't continue to advocate for something that was driving away that much potential business.

Veteran analytics expert Avinash Kaushik coined a great term for the gridlock or poor decision making that can occur in this way. He calls it the "HiPPO"—the Highest Paid Person's Opinion.

Steve Jackson, Chief Analytics Officer for Kwantic in Helsinki, Finland, told me a great "HiPPO" story:

"The HiPPO in question wanted to run a €2 million banner advertising campaign for a mobile PDA on thirty high profile media sites.  However, using previous campaign data about cost per click and subsequent conversion rates, I calculated that it would cost around €16,000 to sell one item. Given that the price of the product was €350, I flagged this as a major waste of money—it would actually be cheaper to give away 1,000 PDAs and generate some great buzz with consumers.

Despite my warnings, the HiPPO decided to run the campaigns anyway. But when after a week they only had 13 sales, it became obvious that I was right, and they stopped the ads. They still spent about €200,000 but that's a lot less than €2 million.

Steve added, "The major plus was that the HiPPO started paying a lot more attention to the data after that!"

Sometimes, the culprit in poor decisions is not an individual, but corporate policy. I've seen a number of examples where a parent company makes rules for subsidiaries located in another country, and often with a different culture.

A recent audience member who works for "a unit of a leading global manufacturer" responded to a survey question about the most successful aspect of their web presence. He said sadly, "N/A. Website managed and maintained at Corporate." Then, he described their greatest challenge in this area as, "Timely forwarding of website information requests by Corporate."

Indeed, when I checked their website, no contact details were provided at all—there was only an online form. Submission of this produced the highly impersonal message: "Thank you for contacting us. We will forward your question/request to the appropriate person, and respond to you as soon as we can." This certainly doesn't give the impression that much value is placed on the visitor's needs.

Surely it would be worth investigating how many of these forms are completed, and how long it takes to forward them appropriately. It also might be eye-opening to the corporate executives to have some idea of the potential business lost through this convoluted procedure. 

Of course, I'm being very tongue in cheek here—business owners can make excellent decisions in short timeframes. The key to well-informed decision making lies in the use of objective facts, intelligence, and hands-on experience, rather than subjective desires.

© 2011 by Philippa Gamse, CMC.