Blog

/

Performance Management

Why Performance Management Has Been So Hard to Get Right

Performance management asks one process to do conflicting jobs with measurements that are less reliable than they look. Here is where reviews break down in practice.

Sarah Katherine Schmidt

November 18, 2026

Illustration of a tilted balance scale beside a stack of blank review forms and a ruler

Performance management has been hard to get right because one process is asked to do jobs that pull against each other, using measurements that are less reliable than they look, on work that no longer holds still.

This is Part 1 of a three-part series on performance management. Part 2 covers where things stand now, and Part 3 covers what HR can do differently.

One conversation, too many jobs

The same review is expected to set pay, decide promotions, build a legal record, and help someone improve. Those purposes conflict. If a rating affects your bonus, you will not tell your manager what you are struggling with. If the goal is development, the conversation cannot also be sorting people into buckets. Most systems try to do both and end up doing neither well. Cappelli and Tavis made this case in their 2016 Harvard Business Review article on the performance management revolution, which covers why Adobe, Deloitte, and Accenture dropped annual ratings. The research likely still holds true but with some caveats.

Ratings say more about the rater

Research by Scullen, Mount, and Goff (2000) found that most of the variation in ratings came from the rater, and a much smaller share reflected actual performance. A rating tells you more about the manager than about the person being rated. Organizations then feed those numbers into pay and layoff decisions as if they were precise.

Then we wonder why employee engagement is the lowest its been in 15 years. Gallup found 29% of U.S. employees engaged in the third quarter of 2026, the lowest since 2011. The Gallup piece does not blame performance management. It points to fewer employees strongly agreeing they know what is expected of them (46%, down from 49%) and that someone at work cares about them (37%, down from 40%). The employee experience is inherently complex.

Whether someone feels engaged depends on clear expectations, a manager who cares, and support through change, and a single rating captures little of that. Ratings also assume that one person's output can be separated from the system around it, which is the next problem.

Individual output is hard to separate from the system

A salesperson with a quota is one case. Most knowledge work depends on handoffs, changing priorities, and decisions made several levels up. When results move, nobody can say how much was one person's doing, so organizations count what is visible: hours online, messages sent, tickets closed. Once those become the target, people optimize for them.

Where reviews break down in practice

The failures follow the review cycle.

  • Goal setting. Goals written in January are often out of date by March, so the review measures people against a plan the business abandoned.
  • Evidence. Managers often rely on memory. Most have a few weeks of clear recall and a general impression of everything before that, so ratings lean toward what happened recently and what was loud.
  • Visibility. The person who presents to leadership and posts updates is easy to remember. The person who fixed a process that was costing the team ten hours a week is not. Hybrid work made this worse, because fewer people are observable.
  • Self-reviews and peer input. Self-reviews favor people who are comfortable describing their own work in strong terms. Peer feedback tends to track who has worked with whom and who is well liked.
  • Calibration. Managers defend their people against a fixed budget or distribution, and outcomes can depend on who argues best.
  • Delivery. The conversation happens once, often weeks after the work, with a number attached.

The result is a rating that reflects the manager's memory, the manager's skill in a meeting, and the employee's visibility. It reflects the employee's contribution only in part, and nobody inside the process can tell how much.

Manager training helps at the edges. It does not solve the underlying problem of a trained, motivated, and rushed manager rating six to ten people on work they only partly saw. Part 2 looks at what organizations are doing about that problem now.

More practical guides

Templates, checklists and cycle calendars in the knowledge hub, free.

Open the knowledge hub
Get started

Turn “we should fix performance” into a repeatable rhythm.

A 30-minute walkthrough with someone who has set this up for a two-person HR team before. No deck.

Price per year$3,000Up to 100 people, with every feature included.
Talk to an HR expertSee pricing