The Hidden Tax Nobody Talks About

Why Performance Reviews Are Quietly Bleeding Your Company Dry

TL;DR Version – Performance Reviews Are Quietly Killing Productivity

• Quarterly review slump is real: A fintech saw a 25% productivity dip during review periods. Managers spent 40% of their time writing docs; employees stopped working to write self-assessments.

• It’s a massive hidden tax:

  • For a 250-person company, reviews can burn 9,600+ manager hours a year

  • Equivalent to £4–5 million in lost opportunity

  • Managers spend 210 hours/year on reviews; employees ~40 hours; HR dedicates 12–15% of their capacity

• You’re paying for failure:

  • Performance reviews cost €1–1.5 million annually for mid-sized orgs

  • Only 15% of employees find them helpful (Gallup)

  • 1/3 of reviews actually decrease performance

• The human cost is worse:

  • Stress, anxiety, and distrust spike around review time

  • Feedback gets bundled with compensation and development, which overloads the brain and makes all of it less effective

• Transparency ≠ Solution:

  • Self-documentation sounds good, but it creates more documentation, more review cycles, more meetings

  • Still subjective, still slow, still inefficient

• We’re stuck using a WWII-era model:

  • Designed for static jobs in a stable economy

  • Totally incompatible with fast-moving, collaborative, knowledge-based work

• AI agents offer a real alternative:

  • Can monitor contributions, spot patterns, and trigger feedback in real-time

  • Separate and optimize performance, development, and compensation conversations

  • Eliminate admin burden without losing human judgment where it matters

Big idea: Performance reviews aren’t just broken—they’re actively hurting your company. AI agents could radically reduce the cost while making feedback actually useful.

Full Version

Well, I know it looks like a repetitive pattern, but I’ll start with the story again, cos I’ve been on this anthropological expedition since March.

So, a friend who leads operations at a 180-person fintech told me this story just two weeks ago that honestly made me question everything I thought I knew about organizational efficiency (No, it’s not ANNA, we are on the opposite side of the spectrum, but it is a European-based entity, can’t reveal my sources, sorry). I was looking for somebody who does reviews as often and as correctly as possible. The leadership team was completely puzzled by this quarterly productivity slump that seemed to happen like clockwork. I mean, every quarter, same story.

After digging into their operations (and probably spending way too much time on spreadsheets), they discovered something that shouldn't have surprised anyone but somehow did: productivity consistently dropped by roughly 25% during the three weeks surrounding their quarterly review process. Managers were spending nearly 40% of their time on review documentation, while employees were anxiously preparing self-assessments instead of, you know, actually doing their jobs.

And I'm sitting there thinking - wait, this can't be normal, right? But then I started looking into it more, and honestly, it's worse than I thought.

Look, I'm not pretending I'm an HR expert here. I'm a product management guy who's been in software process management for years. I got curious about this whole thing because it seemed like such an obvious waste. And I’m using simple common sense and a bit of the power of scientific research.

But here's what I discovered: we're all sitting on this massive hidden cost that nobody really talks about, and it's quietly destroying productivity across organizations.

The Numbers That Made Me Go "Wait, What?"

Companies with 10,000 employees spend up to $35 million annually on performance reviews. I read that and thought it had to be a typo. But even smaller organizations - and I mean much smaller - face proportionally significant costs that most have never even calculated.

The traditional review process¹ creates what I'm calling a "hidden tax" on organizations in three areas: time, money, and morale. And honestly, the more I dig into this, the more frustrated I get.

According to Deloitte research, managers spend an average of 210 hours annually on performance management activities when using conventional systems. That's like... five full work weeks. Just on reviews. Employees spend about 40 hours per year on this stuff. HR teams? They're dedicating roughly 12-15% of their total capacity to administering, troubleshooting, and reporting on these processes.

Here's where those hours actually go:

• About 30% on preparing formal assessments

• Roughly 25% on gathering input and documentation

• Nearly 25% on calibration and review meetings

• The remaining 20% on feedback delivery and follow-up discussions

Let's say we have a 250-person financial services company somewhere in London (I'm just picking cities here to make the math relatable). Their review process might consume approximately 9,600 manager hours annually. Those same hours could generate an estimated £4-5 million in revenue-generating activities if people were actually focusing on client engagement and product development instead of filling out forms.

Cal Newport (you know, the productivity guy) has this great quote: "Administrative overhead is the silent killer of knowledge work value. Every hour spent on documentation is an hour not spent on innovation." And I keep coming back to this because it's so true.

The Real Financial Impact (Spoiler: It's Depressing)

So here's a simple formula I came up with to calculate what this actually costs:

Annual Review Cost = (Average Manager Hourly Rate × Manager Hours) + (Average Employee Hourly Rate × Employee Hours) + (HR Support Costs) + (Technology Costs) + (Opportunity Cost of Diverted Attention)

Let's say we have a 300-person technology company in Stockholm (again, just picking random cities and adjusting for local economy and salaries). This calculation might reveal an annual review cost of approximately €1-1.5 million, or roughly €3,000-5,000 per employee. Most leadership teams are completely stunned when they see these numbers because they've never actually calculated the full financial impact.

Gallup research confirms this is typical, finding that traditional performance reviews cost large companies millions with questionable returns. And get this - their data shows that only about 15% of employees strongly agree that performance reviews inspire improved performance. So we're spending millions for a 15% satisfaction rate. That's... not great.

The indirect costs are equally concerning. Let's say we have a 150-person SaaS company in Amsterdam. They might see roughly a 15-20% decrease in code commits and about a 25% increase in bug rates during the two weeks surrounding their semi-annual review period. This productivity dip alone could cost them an estimated €300,000-400,000 annually in delayed releases and quality issues.

The Human Cost (Which Might Be Even Worse)

Beyond time and money, there's this psychological burden that traditional reviews create, and it's honestly pretty damaging to organizational culture.

According to Gallup, about 95% of managers are dissatisfied with their organization's review systems. Even more concerning - roughly one-third of traditional reviews actually decrease performance rather than improving it. Like, we're actively making things worse.

Josh Bersin (employee experience expert) puts it perfectly: "Annual reviews create artificial anxiety spikes that damage trust. They're the corporate equivalent of final exams, which we know are among the least effective ways to promote learning."

The Transparency Trap (Or Why "Just Do Continuous Feedback" Doesn't Really Work)

I had this interesting conversation on LinkedIn with my friend Alex Gornik about this whole thing. His first thesis was about open and self directed feedback culture via self documentation and ability to other any time add comments and feedback to this documentation. Still, he mentioned that even with public, transparent feedback systems, "many of your questions are answered only by the manager and are subjective to the manager's decision."

And that got me thinking - even when we try to fix this with transparency and self-documentation, we're still missing something fundamental.

Here's the thing: if everyone starts spending time documenting their achievements properly, and then we need validation processes where someone has to review all that documentation, and then provide meaningful feedback for improvement or growth... we haven't actually solved the problem. We've just spread the cost more evenly throughout the year.

On top of that.

The NeuroLeadership Institute research (quite old research but fundamental imo) shows we need to separate three distinct conversations that traditional reviews awkwardly try to combine:

1. Performance/Growth feedback (focused on specific work outputs)

2. Development planning (focused on growth and learning)

3. Compensation discussions (focused on market value and internal equity)

Most of the time, we do join all three of these things together in one go, to save time (We all know, actually, everybody treats performance review as a trigger to get a raise, right? )

David Rock explains it well: "The human brain cannot effectively process feedback, plan development, and discuss compensation simultaneously. These trigger different neural networks that often conflict with each other."

But even when we separate these conversations, we still have a huge overhead for coordination and documentation. And that's when I started thinking... what if we're approaching this entirely wrong? Honestly, the first thing that comes to mind is, "F**k it, I don't need these reviews at all."(here is where ANNA stands for now…ha-ha) As my friend Robin Black said, “Take me for a drink if you *really* know what you’re talking about and care about my well-being." And then I start thinking we're going about it completely wrong. Sorry, Robin, I’m not sure I’ll grab a pint whenever I need to give you feedback. I want to keep the secrecy of pub crawling with colleagues at the level of weather and weekend plans discussions.

---

Here's Where It Gets Interesting (The AI Agent Possibility)

So I've been diving deep into AI and agentic frameworks lately (it's part of my whole educational journey I mentioned), and I keep coming back to this question: what if instead of optimizing human-driven processes, we completely reimagined performance feedback with AI agents?

Think about it for a second. An AI HR agent could continuously observe and document performance patterns without creating this massive time tax on humans. It could provide real-time, potentially less biased feedback based on actual work outputs rather than these memory-based annual assessments that we're all struggling with.

Imagine an AI agent that could:

• Automatically capture and contextualize daily work contributions from tools you're already using

• Identify patterns that humans honestly just miss in productivity, collaboration, and skill development

• Handle those three separate conversations (performance, development, compensation) naturally throughout the year based on actual data triggers

• Generate personalized development suggestions based on real performance gaps, not generic templates

• Eliminate most of the documentation burden while actually increasing feedback accuracy

This isn't about replacing human judgment in complex situations - I'm not advocating for robot overlords here. It's about removing the administrative overhead that's crushing both managers and employees while providing better, more continuous insights.

---

This is part of my ongoing series where I'm basically documenting my journey into understanding AI employees and how they'll change organizations. Next up: Why Data-Driven Feedback Actually Matters (And Why Humans Suck at It, at least for now)

P.S. This post was created with about 80% natural intelligence and a dash of AI assistance. My exploration into AI productivity tools is already paying off, though my fingers still did most of the typing, and my brain definitely did all the scattered thinking.

1. I refer to the current 360performance review procedures when addressing Traditional Performance review, system, etc.