The Thunderbird survey looked at companies which had ‘scandals’ or publicly embarrassing episodes and compared their admiration level to a well known sector peer. It revealed some insights about the long term effect of scandal, and the right and wrong way to go about restoring reputation.
The things companies do to improve reputation are like ‘entries’ in a giant log called the reputation database. As new entries are made, old ones are forgotten because there are so many competing data streams and the RAM is insufficient. Therefore, what is sensational gets recalled.
Information overload is one of the main reasons companies find it difficult to change perception after a traumatic incident like United’s Dr. Daor eviction or Facebook’s Russia hacking.
“Incompetent crisis strategies cause companies to fall into a corporate version of Newton’s Law, which says once a company gets in trouble it tends to stay in trouble.” (click to tweet)
But is there a magic elixir that can improve a damaged corporate reputation?
In the course of our fieldwork we discovered the primary reasons people admire companies (80%) are the qualities of integrity and vision. To restore reputation, it is absolutely necessary to restore those qualities first.
- The widest ‘scandal effect’ was a 60% variance between Google and Facebook after its Cambridge Analytica dustup, followed by an unimpressive performance by Mark Zuckerberg in the Senate hearing; secondly, United’s scandal effect resulted in a 64% variance between it and Southwest ) after Dr. Dao’s disturbing eviction.
- After Uber suffered from turmoil in the C suite, its score is 44% compared to Lyft with 78%.
- BMW scored 81% compared to Volkswagen at 36% after the company’s emission cheating failure.
- Another challenge with reputation failures is the “Nixon Effect.” It refers to the long term damage to reputation that occurs when incompetent crisis management becomes the real crisis.
Scandal v. Admired
(GlobalRank™ rating is a measure of a company’s admiration level. For full methodology, click here)