Article
The Incentive Was Not to Be Right
Compensation, promotion, and recognition systems routinely reward visible activity, volume produced, targets hit, accounts opened, because activity is easy to count, while the harder, quieter judgment of deciding correctly under uncertainty leaves no comparable trail and often goes unrewarded. Peopl
- Published
- July 23, 2026
- Updated
- August 19, 2026
- Reading time
- 9 min
The Global Signal
In September 2016, the Consumer Financial Protection Bureau announced a $185 million settlement with Wells Fargo, the largest such penalty the agency had issued to that point, after finding that Wells Fargo employees had opened deposit and credit card accounts without customers' knowledge or consent, in order to meet aggressive internal cross-selling sales targets tied to compensation and continued employment (CFPB, September 2016 consent order). Wells Fargo's own subsequent internal review found the practice had produced millions of unauthorized accounts over the preceding several years, and the bank ultimately terminated approximately 5,300 employees connected to the practice. In February 2020, the US Department of Justice and the Securities and Exchange Commission announced a further $3 billion settlement with Wells Fargo related to the same conduct (DOJ, February 2020 settlement announcement).
The employees involved were not, in the aggregate, uniquely dishonest people. They were working inside a system that measured and rewarded accounts opened, and did not measure or reward whether the accounts a customer actually needed were the ones being opened.
DOJ and SEC settlement Wells Fargo paid in 2020 related to its sales-practice conduct
A documented settlement figure specific to this case, on top of the CFPB's separate $185 million 2016 settlement; not a general statistic about incentive design.
People are not making bad decisions. They are making the decisions the incentive system actually pays for.
What changes when judgment is measured alongside activity
A volume-based incentive cannot distinguish a correct decision from one that merely produces the counted output faster.
Pairing every volume metric with a separate, independently weighted review of decision correctness closes the gap an incentive system alone cannot see.
Why the Visible Metric Misleads
A volume-based incentive metric, accounts opened, tickets closed, deals signed, measures something real and countable, but it cannot distinguish activity that reflects sound judgment from activity that merely produces the counted result faster. The more revealing measure sits one layer deeper: a sampled, independent review of whether the underlying decision was actually correct for the customer or situation, not simply whether an output was produced. Wells Fargo's own aggressive cross-selling targets measured accounts opened with precision; nothing in that measurement system was built to notice, in real time, whether the accounts being opened were ones customers had actually asked for or needed.
The Leadership Move
The right move is not to eliminate performance metrics tied to compensation. It is to pair every volume-based incentive with a separate, sampled, independent check of decision quality, correctness of the underlying judgment, not just the count of output produced, and to weight that check meaningfully in how people are actually rewarded.
- Ownership
Sales, operations, and frontline management typically own the volume metric and the compensation structure built around it. Risk, quality, or an independent audit function needs to own the separate check of whether the underlying decisions were actually correct, since the team whose compensation depends on the volume metric is poorly positioned to flag that the metric itself may be rewarding the wrong behavior.
- Tradeoff
Building and weighting an independent decision-quality check costs real money and slows down how quickly volume-based incentives can be scaled across a large organization. Wells Fargo's case shows the alternative cost: $185 million in the initial 2016 settlement, a further $3 billion in the 2020 DOJ and SEC settlement, and the termination of roughly 5,300 employees whose behavior the incentive system had, in effect, produced.
- Human consequence
Customers whose credit scores were affected by unauthorized accounts, and the frontline employees who worked inside targets that made unauthorized account-opening a rational response to the incentives they were actually given, both experienced the consequence of a system that measured activity and never measured whether that activity was the right decision.
Implication for Operators
Any organization with a compensation or recognition system tied to a countable, visible activity should assume that activity, not judgment quality, is what the system is actually optimizing for, unless a separate, independently weighted check of decision correctness exists alongside it. The practical shift is building that independent check before scaling a volume-based incentive further, not after a regulator or auditor finds the gap.
Wells Fargo's incentive system was extremely effective at producing what it measured: accounts opened. It was never built to measure, or reward, whether those accounts reflected a correct judgment about what the customer actually needed. The organization did not fail to notice a problem; its own measurement system was incapable of seeing one, because it was never designed to check for it.
The decision blindness here is not employees choosing to do the wrong thing. It is an incentive system built to reward a countable output, with no parallel system built to reward, or even check, whether the underlying decision was right.
How well do your incentives reward correct judgment under uncertainty, not just visible activity like deals closed, tickets resolved, or features shipped?
FAQ
Were the Wells Fargo employees who opened unauthorized accounts uniquely dishonest?
The scale and duration of the practice, millions of accounts over several years according to Wells Fargo's own subsequent review, points to a systemic incentive problem rather than isolated individual dishonesty. The CFPB's 2016 order and the 2020 DOJ and SEC settlement both focus on the sales-practice and compensation structure that produced the behavior at scale.
Is measuring visible activity necessary for managing any large team?
Yes, and this article does not argue against measuring activity. It argues that activity measurement alone, without a separate, independently weighted check of whether the underlying decisions were correct, will reliably produce more activity regardless of its quality.
How can an organization tell if its incentives reward activity over correct judgment?
Ask whether anyone independent of the team being measured samples and reviews whether the underlying decisions behind the counted activity were actually correct, and whether that review meaningfully affects compensation. If the volume metric is the only thing tied to reward, the incentive is for activity, not judgment.
Would adding a decision-quality check just slow down high-performing teams?
It adds real cost, but a sampled, proportionate review does not require reviewing every transaction, only enough to catch a systemic pattern before it reaches the scale Wells Fargo's case did.
Who should own the independent decision-quality review?
A function structurally separate from the team whose compensation depends on the volume metric, typically risk, internal audit, or a customer-outcomes function reporting outside the sales or operations chain being measured.
Related intelligence
Article
Steering From Lagging Indicators
Strategic choices are routinely built on the most recent quarter's or year's results, indicators that are, by construction, a record of conditions that have already changed by the time they are reported. Leadership can steer confidently, using genuinely accurate numbers, toward a destination defined
Article
Automating a Decision Nobody Had Actually Made
Organizations frequently automate a workflow, a trading rule, a routing decision, a pricing exception, before anyone has explicitly decided what that workflow should optimize for, what should happen at its edges, or who owns the exceptions it will inevitably produce. Automation does not remove a dec
Article
Governance as a Way Not to Decide
A review committee, a steering group, or a multi-stage sign-off chain can look like careful oversight while actually functioning as a way to defer an uncomfortable decision indefinitely, because each additional review step looks rigorous and simultaneously postpones the moment someone has to actuall