---
title: "Competency and Culture (video series)"
description: "A 28-part series on building a performance and culture system that holds up: why ratings reflect the rater more than the person, how calibration fixes it, why pay and growth belong in separate conversations, how to hire for culture add instead of “fit,” and how to build a competency framework your managers actually remember. Hosted by Deepti Gupta, co-founder of Sageo."
url: "https://sageo.ai/videos/competency-and-culture"
updated: "2026-10-05T00:00:00.000Z"
source: "Sageo (sageo.ai)"
---


# Competency and Culture

_A 28-part series on building a performance and culture system that holds up: why ratings reflect the rater more than the person, how calibration fixes it, why pay and growth belong in separate conversations, how to hire for culture add instead of “fit,” and how to build a competency framework your managers actually remember. Hosted by Deepti Gupta, co-founder of Sageo._

Full playlist: [Competency and Culture on YouTube](https://www.youtube.com/playlist?list=PLRVszqr29fMY)

## Part 1: Your Performance Ratings Measure the Manager, Not the Person

In 2000, Scullen, Mount and Goff studied more than 4,400 managers, each rated from several perspectives. About 62% of a rating reflected the rater's own standards and habits, and only about 21% the person being rated. A rating is roughly three times more about who does the rating than who receives it.

Watch: [Your Performance Ratings Measure the Manager, Not the Person](https://www.youtube.com/watch?v=pvm16qogAR4)

## Part 2: The Strict Manager and the Distant Manager: Two Wrong Ratings

Two common manager types quietly distort ratings. The strict manager's 3 is everyone else's 4. The distant manager rates what they see in meetings, so strong presenters drift up and the quiet people who keep things running drift down. You only catch either one by comparing ratings across perspectives and across managers.

Watch: [The Strict Manager and the Distant Manager: Two Wrong Ratings](https://www.youtube.com/watch?v=MHCcF6h2ne8)

## Part 3: 5 Biases Hiding in Every Performance Rating

Experienced managers don't rate more fairly, they just get more consistent at the same mistakes. This is the plain-language tour of the five biases in almost every review cycle, what each one costs, and why structure beats experience. It ends with the Deloitte idea of asking about your own intentions rather than judging someone's qualities.

Watch: [5 Biases Hiding in Every Performance Rating](https://www.youtube.com/watch?v=zFf0-XVN_9k)

## Part 4: Calibration in 2 Minutes: A Second Look Before Ratings Count

Calibration is a structured second look at ratings before they turn into pay and promotion decisions. It surfaces manager patterns, disagreement between perspectives, and ratings with no evidence behind them. The one trap to avoid: starting from a quota instead of from the evidence.

Watch: [Calibration in 2 Minutes: A Second Look Before Ratings Count](https://www.youtube.com/watch?v=oXZJm8ukriM)

## Part 5: Your Best Performer Might Be Your Most Expensive Hire

Your best performer can be your most expensive hire, not for what you pay them but for what they cost everyone around them. MIT Sloan found toxic culture predicts attrition far more than pay; Harvard research found avoiding one toxic worker is worth more than twice as much as hiring a superstar. And toxic workers are often more productive, which is exactly why companies keep them.

Watch: [Your Best Performer Might Be Your Most Expensive Hire](https://www.youtube.com/watch?v=VLQbMo-B97o)

## Part 6: Demanding Boss or Brilliant Jerk? How to Tell the Difference

Not every tough boss is a jerk, and punishing demanding leaders is a costly mistake. The simplest line: demanding people raise the standard of the work; brilliant jerks raise the cost of being near them. Six ways to tell them apart, and the one test that cuts through, which is to watch how someone treats people who can't help their career.

Watch: [Demanding Boss or Brilliant Jerk? How to Tell the Difference](https://www.youtube.com/watch?v=3NEcTBXzye4)

## Part 7: How to Spot a Brilliant Jerk Before (and After) You Hire Them

The cheapest time to deal with a brilliant jerk is before you hire them. Five practical checks for your hiring process, from listening for “I” versus “we” to taking references sideways and below, plus five warning signals once someone is on the team. One rule for reading them: one signal is noise, three is a pattern.

Watch: [How to Spot a Brilliant Jerk Before (and After) You Hire Them](https://www.youtube.com/watch?v=ephf0r5tqW0)

## Part 8: Score the What and the How: Fixing Results-Only Reviews

If you only score what people deliver, you keep promoting people who are quietly breaking your team. Score outcomes and behaviour separately and you get four boxes, including the highest-risk one: the toxic producer. Two rules make it work, and one question keeps you honest: if this person left tomorrow, what would actually break, and for how long?

Watch: [Score the What and the How: Fixing Results-Only Reviews](https://www.youtube.com/watch?v=nDs29G9HikU)

## Part 9: Your Company Values Are Just Words (Until Reviews Use Them)

The Culture 500 research found no correlation between the values companies publish and how well employees say they live them. The usual suspects, integrity, collaboration, customer focus and respect, sound the same everywhere and mean something different to everyone. A word is not an instruction, so the real test of a value is whether your review system can actually see it.

Watch: [Your Company Values Are Just Words (Until Reviews Use Them)](https://www.youtube.com/watch?v=igZHN_6QLH0)

## Part 10: What Boeing Teaches Every Company About Values

Every aircraft maker says safety comes first. After the 2024 door-plug incident, an FAA-convened panel reported a disconnect on safety culture and employees hesitant to raise concerns. The lesson for any company: a value only shapes behaviour when the review system both rewards it and protects the people who act on it.

Watch: [What Boeing Teaches Every Company About Values](https://www.youtube.com/watch?v=IQGGALk5ndA)

## Part 11: What Ownership Actually Looks Like at Every Level

Almost every company values ownership; almost none can say what it looks like for a junior engineer versus a head of function. See ownership written as observable behaviour at three levels, with a clear line for below the bar, at the bar and role model, and watch how the bar rises with the role.

Watch: [What Ownership Actually Looks Like at Every Level](https://www.youtube.com/watch?v=Ii9fvqTZ7aM)

## Part 12: How to Turn a Company Value Into Observable Behaviour

Don't write your values in a workshop; find them in the moments that already happened. A four-step method, the critical incident technique: collect real moments, pull out the behaviour, sort by level and standard, and test the definitions with a second group of managers. If two groups can't agree what a behaviour means, neither will a real review.

Watch: [How to Turn a Company Value Into Observable Behaviour](https://www.youtube.com/watch?v=MbknddhLa0E)

## Part 13: “Not a Culture Fit”: The Most Expensive Sentence in Hiring

“Not a culture fit” may be the most expensive sentence in hiring, because most of the time nobody can say what it means. A 2025 study of recorded interviews found fit judged on conversational style, not ability, and Lauren Rivera's work found elite firms defining fit as similarity. When fit isn't defined, comfort fills the gap, and comfort looks like the person doing the hiring.

Watch: [“Not a Culture Fit”: The Most Expensive Sentence in Hiring](https://www.youtube.com/watch?v=hpGjVLCbl4g)

## Part 14: The Product Manager Who “Wasn't a Fit” (And What Happened Next)

A product manager was rejected as “not a fit” for questioning decisions and pushing back. Two years later, the same behaviours made them a function lead somewhere that valued debate. The iceberg model explains why: fit judgments read style, above the waterline, not the values below it. Before you reject someone, ask whether you're reacting to their values or just their style.

Watch: [The Product Manager Who “Wasn't a Fit” (And What Happened Next)](https://www.youtube.com/watch?v=bZ2hzq1er9k)

## Part 15: Culture Fit vs Culture Add: Stop Hiring People Like You

Culture fit and culture add are not the same thing. Fit rewards familiarity and filters out people who are different; add rewards contribution and filters out people who don't share your values. Culture add still has a bar, just the right bar, and it only works if your values are written as real behaviour.

Watch: [Culture Fit vs Culture Add: Stop Hiring People Like You](https://www.youtube.com/watch?v=fgSuJfPjnaA)

## Part 16: 4 Interview Changes to Stop Hiring for Familiarity

You don't need a new hiring process to stop hiring for familiarity. Four changes: retire “fit” as a verdict, test values like a skill, ask what each person would add, and separate style from substance. If you only do one thing, add one mandatory scorecard line: what would this person add?

Watch: [4 Interview Changes to Stop Hiring for Familiarity](https://www.youtube.com/watch?v=SfzEQ-wGtw4)

## Part 17: Why Nobody Remembers Your Feedback (Once Pay Enters the Room)

Spend twenty minutes on thoughtful feedback, then tell someone their raise, and only the number lands. Kluger and DeNisi's 1996 meta-analysis found feedback made performance worse in over a third of cases, especially when it points at the self. A salary number is the loudest self-signal there is, so growth and pay belong in separate conversations.

Watch: [Why Nobody Remembers Your Feedback (Once Pay Enters the Room)](https://www.youtube.com/watch?v=APewWRd1dek)

## Part 18: Only 2% of HR Leaders Say Their Reviews Make People Better

Gallup found just 2% of Fortune 500 chief HR officers strongly agree their performance system inspires employees to improve. What does work is regular, growth-only progress conversations, which Gallup links to far higher engagement and a fairer-feeling process. The fix isn't a better form, it's more conversations, more often, kept separate from pay.

Watch: [Only 2% of HR Leaders Say Their Reviews Make People Better](https://www.youtube.com/watch?v=kWQl9-zdMRg)

## Part 19: Separate Pay and Performance Reviews Without Breaking the Link

Taking pay out of the review doesn't break the link between performance and pay; it changes the order and the forum. A four-step sequence, develop, calibrate, decide, tell, with the growth and pay conversations weeks apart. Google ran a version of this, described in Laszlo Bock's Work Rules, because people can't absorb feedback while waiting to hear their number.

Watch: [Separate Pay and Performance Reviews Without Breaking the Link](https://www.youtube.com/watch?v=AKycxBtEkTU)

## Part 20: How to Run a Growth Conversation: SBI and GROW Explained

Take the number out of the room and you need a real growth conversation. Two frameworks make it easy: SBI (Situation, Behaviour, Impact) for feedback on what happened, and GROW (Goal, Reality, Options, Way forward) for planning what's next. A pay conversation asks what you're worth; a development conversation asks how you get better. Asked in the same hour, only the first gets heard.

Watch: [How to Run a Growth Conversation: SBI and GROW Explained](https://www.youtube.com/watch?v=H1R0SuKC414)

## Part 21: Why 360 Reviews Don't Fix Biased Ratings

When one manager's rating isn't reliable, the instinct is to add raters. But about 62% of a rating reflects the rater, not the person (Scullen, Mount and Goff), and adding raters adds rulers, not accuracy. Averaging six biased views gives you a smoother number, not a truer one. The fix is a shared standard and a room where ratings meet evidence: calibration.

Watch: [Why 360 Reviews Don't Fix Biased Ratings](https://www.youtube.com/watch?v=6G8Ib4KmN18)

## Part 22: What Zalando's Employee Rating System Teaches Every Founder

Zalando's Zonar let employees rate each other continuously, with an algorithm sorting results into tiers that drove pay. A 2019 Hans-Böckler-Stiftung study reported stress and, in some teams, only 2 to 3% reaching the top tier; Zalando disputed it. The lesson holds either way: more ratings with no human evidence check, wired to money, scales bias rather than removing it.

Watch: [What Zalando's Employee Rating System Teaches Every Founder](https://www.youtube.com/watch?v=QJcKFL2jDNI)

## Part 23: How to Run a Calibration Session: 5 Steps, One Rule

A calibration session runs on one rule: no score without evidence. Five steps, prepare, compare by team, discuss the gaps, test the evidence, record the reason, plus what's deliberately not on the list, a target distribution. Two real cases show it catching a strict manager's low rating and a skip-level's inflated one.

Watch: [How to Run a Calibration Session: 5 Steps, One Rule](https://www.youtube.com/watch?v=BNUjoQxHUAQ)

## Part 24: Calibration vs Forced Ranking: Where Is Your Line?

Calibration becomes forced ranking the moment it starts from a quota instead of evidence, as in Meta's reported 2025 instruction to place 15 to 20% of people in “below expectations.” A side-by-side of the two, and one clarification: ratings still go down in calibration, the difference is that the outcome isn't decided in advance.

Watch: [Calibration vs Forced Ranking: Where Is Your Line?](https://www.youtube.com/watch?v=TcWN6cqKKxE)

## Part 25: The 30-Second Test for Your Competency Framework

Run this test: ask a manager to name the competencies they rate their team on, from memory. Most can't, and if they can't name them, they aren't using them. Frameworks grow one reasonable request at a time until managers tune out, priorities blur and calibration breaks. If your managers fail the test, the problem isn't your managers, it's the list.

Watch: [The 30-Second Test for Your Competency Framework](https://www.youtube.com/watch?v=F8A8AVb9_XE)

## Part 26: Why Long Competency Lists Make Reviews Less Fair

A long competency list feels thorough but is actually a fairness problem. One manager with eight reports and 25 competencies makes 200 separate ratings a cycle; cut to eight competencies and it is 64. More ratings means more chances for bias to creep in, and it makes evidence-based calibration impossible. Fewer competencies means fewer judgments, better evidence and fairer reviews.

Watch: [Why Long Competency Lists Make Reviews Less Fair](https://www.youtube.com/watch?v=r9og_jT1nOg)

## Part 27: Stop Copying Big Company Competency Frameworks

Most startups don't build a bloated framework, they download one. Enterprise libraries like Korn Ferry's (38 competencies) were built for big organisations with the talent teams you don't have. Use a library as a menu, not a meal. Even Amazon's 16 Leadership Principles work only because they are used every week, not because of the number on the page.

Watch: [Stop Copying Big Company Competency Frameworks](https://www.youtube.com/watch?v=1qFN4TI_drU)

## Part 28: How to Build a Competency Framework With Fewer Than 10 Items

The competency structure for any company under a few hundred people: three layers, company behaviours for everyone, role competencies per role, and a small leadership add-on, usually six to eight per person. Three rules keep it honest: values become behaviours, competencies separate great from good enough, and every competency needs evidence. A short list your managers remember is how culture, measurement and reward line up.

Watch: [How to Build a Competency Framework With Fewer Than 10 Items](https://www.youtube.com/watch?v=32pEXWFc5x0)