We’re at it again. Wells Fargo is being fined $185,000,000 because employees created about 2,000,000 bogus accounts. Why would good people do this? To meet performance metrics, that’s why.
I can hear the management meetings: “Hey, we’re leaving money on the table. Let’s increase our share of wallet. Our average customer has 6.2 of our products. Our goal next year is to get that up to 7.3. Not only will it be good for customers, but you’ll get a bonus.”
The problem is that the Wells Fargo clients are saturated with accounts and don’t need more products. So, how do you hit the goal? Easy, you create accounts with the data at hand, but without asking clients.
Figures lie and liars figure
Creating numbers to satisfy metrics is an old game. My stats professor used to say that “figures lie and liars figure.” Need I remind you of the Veteran’s Affairs scandal two years ago when production numbers were massaged to show that vets really were getting timely service – even though they weren’t.
I learned this game 40 years ago when I was recruiting. My boss taught me how to make our metrics look good by “re-starting” transactions that had gone stale. According to our books, we filled requests in eight weeks while mangers lamented that it took us eight months.
Why do metrics drag us into this cycle of deceit? The short answer is because we’re not as smart as we think we are. Because humans think logically and rationally, we believe that human behavior can be predicted logically and rationally. We believe that by setting performance metrics that we can “socially engineer” human behavior in organizations. Wrong!
Rewards motivate – to get the reward
I spent 10 years in the metrics and performance pay game. It is based on the erroneous assumption that human behavior is predictable. What I learned is that rewards absolutely motivate people, but only to get rewards – and often through aberrant behavior like we’ve seen at Wells Fargo and the VA. Continuous tweaking of the metrics never achieves the desired behavior. If you don’t believe me, listen to Jim Collins in Good To Great:
The idea that the structure of executive compensation is a key driver in corporate performance is simply not supported by the data.
Influencing human behavior is a complex subject that is treated well in Surfing the Edge of Chaos by Richard Pascale where he reminds us that “… we can’t direct a complex adaptive system” and in fact “…efforts to direct living systems beyond very general goals are counterproductive.”
To what metrics do you manage?
Don’t get me wrong. I’m not throwing out metrics. I believe that metrics make jobs tangible by communicating expectations. Where I disagree is when we manage toward short term metrics rather than long term aiming points.
Metrics tell managers how well their organization has performed and where it needs to improve; but answers to complex business challenges aren’t found in the bottom right hand cell of an excel spread sheet. Those answers are found in managerial insight gained from the experience of actually watching people do their work.