Goodhart's Law at Work
When a measure becomes a target, it ceases to be a good measure. Wells Fargo showed how far that goes. The fix is to watch the signals nobody can push on directly.
Goodhart's Law holds that when a measure becomes a target, it ceases to be a good measure. That line is anthropologist Marilyn Strathern's, from 1997. Economist Charles Goodhart's 1975 original was drier: any observed statistical regularity tends to collapse once pressure is placed on it for control purposes.
Wells Fargo ran the experiment at scale. In 2016 the Consumer Financial Protection Bureau found that employees chasing sales targets and bonuses had opened roughly 1.5 million deposit accounts and 565,000 credit card accounts that customers may never have authorized. Every one of them counted toward a target.
The numbers were accurate. They had stopped describing the business, because people could move them directly.
That tells you which signals survive pressure. Revenue, deals closed, accounts opened: each is something a person can push on by itself. Signals further upstream, like whether a team agrees on this week's priorities or whether commitments move between check-ins, are awkward to fake. Faking them means faking the work.
The nudge: Keep your targets. Beside each one, name the upstream signal that should move first if the target is being hit honestly. When the target climbs and that signal stays flat, look closer before you celebrate.
Source: Strathern, 'Improving ratings': audit in the British University system, European Review (1997); Consumer Financial Protection Bureau, Wells Fargo enforcement action (2016)
