By Anne Walker, Founder of LEVEL 110
Most performance management systems are built to manage risk, not accelerate performance. They're designed to minimize employment relations exposure, not to enable the strongest performers in the business to do the best work of their careers. It's time to ask a different question entirely.
For decades, the central question of performance management has been: "How do we document performance well enough to defend our decisions?" Annual reviews. Rating scales. Forced distributions. Stack rankings. Every one of these tools was designed with legal defensibility in mind first and performance acceleration a distant second.
The result is a system that captures what happened twelve months ago, assigns a number to it, and calls that performance management. It isn't. It's performance archaeology.
The right question is entirely different: What does high performance look like for each person in their role, how can we make that measurable, and how can we move from performance as a lagging indicator to a predictive one?
One of the most consequential conversations in any organization happens in a 1:1 meeting between a manager and a direct report. What was agreed to? What did the manager commit to doing? What does the employee need to move faster? What gets revisited next time?
In most companies, those conversations happen and then disappear. Nothing is captured. Nothing is tracked. The manager remembers their version. The employee remembers theirs. Three months later, both parties are operating on different assumptions about what was said and what was expected.
Tools now exist, including AI-powered meeting assistants and conversation intelligence platforms, that can capture what was discussed, what commitments were made, who owns what, and what needs to be revisited. Many companies resist them, citing privacy concerns and discomfort. That discomfort is worth examining. If a manager is uncomfortable having their commitments to their team documented, that is a performance signal in itself.
A performance system that captures real conversations, real commitments, and real follow-through is not surveillance. It's accountability, applied equally to managers and employees.
The 18-month employee survey is one of the most expensive performance management failures in corporate history. By the time the results are analyzed, presented, and acted on, the people who gave the feedback have either moved on, given up, or forgotten they answered the questions.
High-performing organizations have moved to continuous listening: short, frequent pulse checks that surface real-time signals about team health, manager effectiveness, and engagement risk. Not to generate slide decks. To trigger conversations and interventions while there's still time to act.
The difference between a company that finds out about a team in crisis through an exit interview and one that finds out three months earlier through a declining pulse score is the difference between losing the person and keeping them.
Here is the question most performance systems never answer: What does exceptional performance look like in this specific role, at this specific stage of the company?
Not a generic competency framework. Not a five-point scale applied uniformly across every function. A clear, specific, role-level definition of what outstanding looks like, so that every person in the organization knows what they're being measured against, and so that managers can have honest conversations grounded in something concrete.
Without that definition, performance ratings are largely subjective. Managers rate people against their own expectations, which vary wildly. Calibration sessions become debates about individual perceptions rather than assessments against a shared standard. The result is that strong performers in poorly-defined roles get underrated, and average performers with visible managers get overrated.
The good news is that building role-level performance standards no longer has to be the daunting, manual exercise it once was. AI can do the heavy lifting. A manager who understands their team's work can prompt an AI tool to draft performance standards for a specific role, covering what meeting expectations looks like, what exceeding looks like, and what exceptional looks like, in minutes. Their job is not to build it from scratch. Their job is to validate that it's accurate, refine it based on what they know about the role, and keep it current as the role evolves.
That shifts the manager's responsibility from documentation to judgment, which is where their time should be spent. The organization's responsibility is equally clear: hold managers accountable for creating performance standards that are specific, measurable, and actionable. Vague standards produce vague feedback. Vague feedback produces stalled performance and frustrated employees.
This is the question that makes most HR teams uncomfortable, and most CEOs quietly wonder about.
The traditional merit system operates on a curve. A fixed percentage of employees are "top performers," a fixed percentage are "below expectations," and the majority land in the middle. That distribution is assumed, not derived. It exists because someone decided it should, not because it reflects reality.
The curve creates several predictable problems. It forces managers to rank people against each other rather than against a performance standard. It guarantees that a percentage of your workforce will be labeled below expectations even in a year when the company exceeded every goal. It creates resentment, undermines collaboration, and produces compensation decisions that feel arbitrary to the people receiving them.
The alternative is a standard-based system. Every role has a clear definition of what meeting expectations looks like, what exceeding looks like, and what exceptional looks like. People are assessed against that standard, not against each other. In a year where the company hired exceptionally well and developed its people effectively, it is entirely possible that 80% of the organization exceeds expectations. That's not grade inflation. That's what a high-performing organization should look like.
The CEO who is willing to build a standard-based system is making a statement about what they believe about their people. They believe that performance is not a fixed distribution. It's a result of clarity, development, and accountability. And they're right.
Moving from lagging to predictive performance management requires five things most companies have not built:
Written definitions of what meeting, exceeding, and exceptional performance looks like in each role. AI can draft these standards in minutes from a good prompt. The manager's job is to validate, refine, and keep them current. The organization's job is to hold managers accountable for making them specific and actionable, not generic.
A system, whether AI-assisted or structured manually, that documents what was committed to in 1:1 conversations and tracks whether it happened. Applied to managers and employees equally. This is the accountability infrastructure most organizations are missing.
Pulse surveys at a cadence that allows early intervention: monthly or bi-monthly, short enough that people actually complete them, specific enough that the data tells you something actionable. Tied directly to manager accountability conversations, not HR reporting cycles.
The metrics that predict performance problems before they show up in results. Goal completion rate trends, 1:1 frequency and quality, skip-level feedback patterns, internal mobility signals, project delivery consistency. These are the signals that tell you where performance is heading, not where it has been.
The single most powerful lever in any performance system is manager quality. Research is unambiguous: the biggest predictor of team performance is the manager. A predictive system measures manager effectiveness explicitly, through direct report feedback, team retention, goal attainment, and development of talent, and holds managers accountable for those outcomes.
Here is the honest conversation most organizations haven't had.
HR leaders are frequently handed mandates, such as modernizing performance management, implementing AI, or building a predictive system, without the budget, technology partnerships, executive sponsorship, or organizational runway to actually deliver them. When a CEO asks for innovation but doesn't create the conditions for it, the HR team is set up to maintain the status quo, not transform it. That's not a failure of ambition. It's a failure of partnership.
Building a real-time, predictive performance system requires more than HR's effort. It requires the CEO to understand what the transformation actually involves: not just the technology, but the culture shift, the process redesign, the manager enablement, and the time it takes to do it right. It requires the CFO to fund it adequately. It requires engineering and technology leadership in tech companies to engage, not delegate. And it requires every senior leader to model the accountability standards they're asking their teams to meet.
HR cannot do this alone. And in most organizations, they've been asked to. When AI implementation stalls or performance management stays stuck, the question worth asking is not "why isn't HR moving faster?" but "what does HR need from us that they don't currently have?"
The CEO's role is to be an active partner in this work, not a sponsor who approves a budget line and waits for results. That means staying close to the implementation, removing barriers when they surface, and being specific about the outcomes that matter: early signals on talent risk, real-time visibility into team health, manager accountability that goes beyond completion rates, and a direct line between individual performance standards and business results.
The performance management system your company deserves is built by a CEO and HR function that trust each other, set clear expectations together, and share accountability for the outcome. That partnership is rarer than it should be. It's also the only thing that makes transformation possible.
Anne Walker is the Founder of LEVEL 110, a San Diego-based executive HR consulting firm. She works directly with CEOs and senior leaders to remove organizational drag, modernize the People function using AI, and build leadership systems that scale.
Take the free Organizational Readiness Check to see where your performance and accountability systems may be creating drag.