# Quarterly Calibration: Score Every Employee on Five Things, Fairly **Speaker:** Patrick Bet-David, with Tom Ellsworth **When:** 2026-09-02 · 5:01 PM–5:29 PM ET (2:01 PM Las Vegas) **Theme:** Build the team **Sources:** OycAYqFAC7qNpvLi2DKB **Notes page:** https://vault.chels.ai/sessions/d2-calibration-framework/ > **Replace gut-feel performance management with a quarterly calibration: every employee self-scores 0-10 on effort, attitude, leadership, innovation and results, the manager scores independently, and a leadership committee reconciles the two so pay, promotion and exits follow data instead of personality.** Patrick Bet-David told the story of running a commission-only insurance sales force on personality for seven years and realising nobody on his team earned six figures. He fixed it with weekly activity accountability, then carried the same logic into salaried companies as a quarterly calibration system built on five scored categories. He explained the score bands, how bonuses multiply for top performers, why employees must rate themselves first, and how the exercise exposes three kinds of managers. Tom Ellsworth added what two years of calibration at Valuetainment revealed about deadlines, turnover and which managers to bet on. Product references to Valuetainment's calibration software are treated here as context; the practices stand on their own. **Best for:** Owners with 25 or more salaried employees who have never run a formal review cycle; Sales leaders managing commission-only or 1099 reps; Founders moving from contractors to W2 staff with salaries, bonuses and equity; Managers who avoid conflict and rate everyone highly; HR and people leaders designing bonus and raise policy; Leaders who suspect a middle manager is hiding or driving out talent ## Claude skills from this session - `quarterly-calibration` — Design and run Patrick Bet-David's quarterly calibration so pay, promotion and exits follow data instead of personality. Scores every employee 0-10 on effort, attitude, leadership, innovation and results (50 points), places them in five bands with fixed consequences (5-10 exit, 11-24 one quarter to fix, 25-39 meets, 40-47 exceeds, 48-50 outstanding), runs the self-score then manager then committee sequence, diagnoses managers from their rating spread (people-pleaser, elitist, steward), and computes bonus multipliers. Use whenever the user mentions performance reviews, annual reviews, bonuses, raises, KPIs, underperformers, a manager who rates everyone highly, a team over about 25 people, or asks how to hold salaried staff accountable, even if they never say "calibration". https://vault.chels.ai/skills/quarterly-calibration/SKILL.md ## Skills to build - **Activity-first accountability (the Monday board)** — Holding people to leading indicators they control, such as calls made and appointments run, before discussing results. Bet-David called his version DMAs, daily monitoring of activity. _How to practise:_ Every Monday, have each rep or team member report last week's activity numbers out loud before any result. Then ask to see the calendar and do the math on hours actually worked. You know it works when reported activity, calendar evidence and results start to agree and 'I work hard' disappears from the conversation. - **Scoring people on five defined categories** — Rating each employee 0-10 on effort, attitude, leadership, innovation and results (EALIR), each defined by about four observable criteria, for a total out of 50. _How to practise:_ Write your own four criteria under each of the five headings using the speaker's list as a starting point. Score a handful of people privately this week and check whether you could defend each number with a specific example. Repeat every quarter; scoring gets faster and disagreements shrink as definitions sharpen. - **Running a calibration committee** — A quarterly meeting where each manager presents scores for their direct reports and other leaders challenge them, asking why someone got an exceeds or a does-not-meet, before scores are final. _How to practise:_ Once a quarter, put managers in a room with their scores. Leave the 'meets' alone unless challenged; debate every 'exceeds' (push up to outstanding or down to meets) and every 'does not meet'. Watch how logical versus emotional each manager's reasoning is. It is working when managers arrive with examples, not adjectives. - **Closing the self-score gap** — Having every employee rate themselves first, then using the gap between their number and the manager's number as the agenda for the review conversation. _How to practise:_ Before each review, collect the self-score. If someone gives themselves 47 and you have them at 31, open with 'tell me why you think you were a 47 last quarter' and go to the numbers. Practice this once per direct report per quarter. Success looks like the gap narrowing to a few points by the next cycle. - **Reading a manager from their rating distribution** — Diagnosing a manager as people-pleaser, elitist or steward by looking at the spread of scores they hand out compared with the score they give themselves. _How to practise:_ After each cycle, chart every manager's distribution across the five bands. Half the team outstanding means conflict avoidance. Manager high, everyone else low means an elitist. A rough 10/20/50/15/5 spread means a steward. Do this every quarter and track whether the same managers drift toward the steward shape. - **Skip-level (overlapping) meetings** — The owner or senior leader talking directly with the people one level below a manager to learn who is really there and whether the manager's scores are believable. _How to practise:_ Pick one manager's team each month and hold short conversations with two or three of their reports. Ask what they are working on, what is in their way and how they get feedback. Compare what you hear with the manager's calibration scores. You are doing it right when hidden strong performers surface before they resign. - **Managing expectations up front** — Setting explicit rules before conflict happens, whether that is a scoring rubric, published salary tiers like Apple's seven engineering levels, or a household 'your money, my money, our money' agreement. _How to practise:_ For any recurring friction (raises, bonuses, who pays for what), write down the rule once, share it with everyone affected, and refer back to it instead of renegotiating. Review the rules annually. It works when the same argument stops recurring. ## What matters - **Personality-based management caps your team's income:** After seven and a half years of leading by charisma, Bet-David earned about 350,000 dollars a year while nobody on his team cleared 100,000. The switch to systems, transfer of knowledge and numbers-based accountability is what let others win too. If your team's ceiling is well below yours, the bottleneck is probably your management style, not their talent. _(Evidence: 17:02:31-17:02:37 / OycAYqFAC7qNpvLi2DKB)_ - **Ask for activity before results:** Results lag and can be blamed on the market. Activity (calls, appointments, hours actually worked) is inside the person's control and is visible on a calendar. Start every accountability conversation there. _(Evidence: 17:03:48 / OycAYqFAC7qNpvLi2DKB)_ - **The calendar exposes the 'I work hard' story:** One rep sat in the office 12 hours a day but ran four one-hour appointments a week, meaning four hours of real work, and earned 1,500 dollars a month. Presence is not activity. Multiply appointments by their duration and you have the true hours worked. _(Evidence: 17:03:48-17:05:09 / OycAYqFAC7qNpvLi2DKB)_ - **Transparent accountability builds trust rather than resentment:** Bet-David found that the harder he drove with visible numbers, the more people appreciated the transparency because they trusted the results it would bring. People resent vague judgement, not clear standards. _(Evidence: 17:06:19-17:07:08 / OycAYqFAC7qNpvLi2DKB)_ - **Salaried teams need calibration more than commission teams do:** Commission-only sales is self-policing: no sale, no pay. Once you pay salary, commissions, 401k and equity and hand people 90 percent of their leads, the market no longer holds them accountable, so you have to build the mechanism yourself. _(Evidence: 17:07:08-17:07:34 / OycAYqFAC7qNpvLi2DKB)_ - **Five categories keep the review broader than one output number:** Effort, attitude, leadership, innovation and results each carry equal weight (10 points). A pure results score misses the person who hits targets while poisoning the team, and the person building future capacity whose numbers lag this quarter. _(Evidence: 17:08:04-17:10:27 / OycAYqFAC7qNpvLi2DKB)_ - **The 'one club' is the clearest marker of a promotable person:** One-club people need to be asked once and it gets done; they bring it back to you when you have forgotten. Bet-David rates these people as priceless and says the more you are in the one club, the faster you move up. Count how many times you have to ask someone and you have a results metric. _(Evidence: 17:09:25-17:10:00 / OycAYqFAC7qNpvLi2DKB)_ - **Make the bottom two bands black and white:** Scores of 5-10 mean the person leaves. Scores of 11-24 get one quarter to fix it and no bonus or raise if they end the year there. Bet-David does not negotiate when someone threatens to leave over a missed raise. Certainty at the bottom is what makes the top feel fair. _(Evidence: 17:10:47-17:11:29 / OycAYqFAC7qNpvLi2DKB)_ - **Multipliers make outstanding worth chasing:** Exceeds (40-47) and outstanding (48-50) earn a multiplier on the target bonus. On a 100,000 dollar base with a 20 percent bonus, outstanding can pay 150 percent of target, or 30,000 instead of 20,000. The accelerator gives ambitious people a number to aim at. _(Evidence: 17:11:34-17:12:03 / OycAYqFAC7qNpvLi2DKB)_ - **People from big companies expect calibration; people from small ones have never seen it:** Bet-David noticed a pattern: hires from billion- and trillion-dollar companies treat calibration as normal, while hires from small firms are unsettled by it. Expect to coach the second group through their first cycle rather than assuming resistance means the system is wrong. _(Evidence: 17:12:50-17:13:37 / OycAYqFAC7qNpvLi2DKB)_ - **A manager who rates nobody low is afraid of conflict:** The people-pleaser hands out 50 percent outstanding, 30 percent exceeds, 20 percent meets and nothing below. Bet-David treats that report as a diagnosis of the manager, not the team, and Tom Ellsworth found those teams were consistently slow on deadlines. _(Evidence: 17:13:56-17:14:45, 17:22:18 / OycAYqFAC7qNpvLi2DKB)_ - **Self-scoring surfaces the expectation gap before it becomes a resignation:** If an employee thinks they are a 47 and you think they are a 32, both sides operate on false assumptions: they believe you are happy, you believe they know you are not. A 15-point disparity is a conflict waiting to happen. Naming it in the review lets you manage expectations and watch the gap close next quarter. _(Evidence: 17:15:55-17:17:16 / OycAYqFAC7qNpvLi2DKB)_ - **The elitist manager costs you your future stars:** A manager who scores themselves outstanding and everyone else poorly makes good people conclude that the company either cannot see the problem or does not care. Two future superstars quit and excel elsewhere. The first calibration exposes this manager, provided you look. _(Evidence: 17:17:45-17:19:13 / OycAYqFAC7qNpvLi2DKB)_ - **Do not teach managers how to calibrate the first time:** Valuetainment deliberately tells managers only 'calibrate your team and come to us'. Watching what they do without coaching reveals their natural bias faster than any training would. _(Evidence: 17:19:31-17:19:44 / OycAYqFAC7qNpvLi2DKB)_ - **The steward's distribution is the company target:** A fair manager lands roughly 10 percent does not meet, 20 percent needs improvement, 50 percent meets, 15 percent exceeds and 5 percent outstanding. Bet-David set that shape as the organisational goal because once everyone sees it, expectations are clear and the whole curve can be pushed upward. _(Evidence: 17:19:44-17:20:45 / OycAYqFAC7qNpvLi2DKB)_ - **Calibration tells you whether the leak is the employee or the manager:** With scores across the whole company you can see whether underperformance clusters under one manager. If it does, invest in that manager. If it does not, coach the employee. Without the data you are guessing about where the money is leaking. _(Evidence: 17:21:18-17:21:42 / OycAYqFAC7qNpvLi2DKB)_ - **Expect about 5 percent to leave when you introduce it:** Two years in, roughly 5 percent of Valuetainment staff exited because they disliked being measured. Those who stayed said they were relieved to be judged on a score rather than background, ethnicity or where they came from. Budget for some attrition and treat it as the system working. _(Evidence: 17:21:45-17:22:18 / OycAYqFAC7qNpvLi2DKB)_ - **Balanced managers are who you call in a crisis:** Tom Ellsworth's review of two years of data found stewards hit deadlines and were the people leadership turned to under pressure, while elitists hit deadlines but had the highest turnover of good people. Calibration exposed which managers to promote to director and which to move out. _(Evidence: 17:22:18-17:23:32 / OycAYqFAC7qNpvLi2DKB)_ - **Fixing the managers is the founder's job, not HR's:** Ellsworth was blunt that moving out weak managers and installing strong ones was the leadership team's responsibility, because the top 20 percent of staff they wanted to keep were yearning for great leadership. _(Evidence: 17:22:46-17:23:32 / OycAYqFAC7qNpvLi2DKB)_ - **Great operators remove steps before they automate:** Bet-David cited The Algorithm, a book by a former Tesla president, on cutting the Tesla checkout from 64 steps to 13 by finding a bank that would do one-page financing. The five-step rule (question, delete, simplify, accelerate, automate last) applies just as well to a bloated review process. _(Evidence: 17:23:32-17:24:49 / OycAYqFAC7qNpvLi2DKB)_ - **Published levels eliminate conflict:** Apple publishes seven engineering tiers, each with a pay band, so a new hire knows exactly what a tier-two engineer must do and what a tier-seven engineer earns. Clarity about the ladder removes most arguments about fairness before they start. _(Evidence: 17:25:09-17:25:48 / OycAYqFAC7qNpvLi2DKB)_ - **Managing expectations is one habit applied everywhere:** Bet-David's 'your money, my money, our money' agreement with his wife has meant zero financial arguments in their marriage. He uses the same instinct with sponsors, attendees and employees. If you are clear in advance, you can afford to be generous afterwards, including life-changing bonuses for people who go above and beyond. _(Evidence: 17:25:48-17:29:06 / OycAYqFAC7qNpvLi2DKB)_ ## Frameworks, lists & numbers - **EALIR: the five calibration categories** — Effort, Attitude, Leadership, Innovation, Results. Each scored 0-10 for a total out of 50. Effort = urgency, quality control (QC), personal development, raising standards. Attitude = perseverance, buy-in, being coachable to the culture, personal character. Leadership = teamwork, developing others, confronting respectfully on these five things, accountability. Innovation = creative and critical thinking, improvement, implementation. Results = being in the 'one club', hitting KPIs, beating your prior best. - **The five score bands and what they trigger** — 5-10 = does not meet expectations: the person is let go. 11-24 = needs improvement: one quarter to clean up, and no bonus or raise if still there at year end. 25-39 = meets expectations: doing a good job, target bonus (band inferred from the stated neighbours). 40-47 = exceeds expectations: bonus multiplier. 48-50 = outstanding: larger multiplier. - **Bonus multiplier example** — 100,000 dollar base with a 20 percent target bonus pays 20,000 at meets. Outstanding can pay 150 percent of the 20 percent, or 30,000. The multiplier is an accelerator so people chase the top band. - **Three manager types by rating distribution** — People-pleaser: 50 percent outstanding, 30 percent exceeds, 20 percent meets, nobody lower; afraid of conflict; teams slow on deadlines. Elitist: rates self exceeds or outstanding, rates everyone else low; hits deadlines but highest turnover of good people; drives out future stars. Steward: 10 percent does not meet, 20 percent needs improvement, 50 percent meets, 15 percent exceeds, 5 percent outstanding; hits deadlines; the person you call in a crisis. The steward shape is the company target. - **Three-view calibration sequence** — 1. Employee self-scores. 2. Manager scores independently. 3. Leadership committee reconciles (example: self 47, manager 39, final 41). Meets scores are generally left alone unless challenged. Every exceeds is debated up to outstanding or down to meets. Every does-not-meet and needs-improvement is questioned. - **The one club** — A person you only have to ask once. Contrast with the two club, four club or seven club, who need repeated reminders. One-club membership is scored under Results and is the fastest route to promotion. - **DMAs: daily monitoring of activity** — Bet-David's 2008-2009 sales accountability system. Monday morning, on the board: calls made last week, appointments run, activity created, before any result. Then 'hand me your calendar' and multiply appointments by duration to compute real hours worked. Example: 238 calls, 28 appointments; another rep's four one-hour appointments equalled four hours of work. - **Calibration cost rule of thumb** — Bet-David's claim: a company with 25 or more employees that does not calibrate quarterly pays five to ten times more in inefficiency than a calibration system would cost. Stated while pitching his own software, so treat the multiple as an opinion, but the 25-employee threshold is a useful trigger for formalising reviews. - **Expected attrition when introducing calibration** — About 5 percent of Valuetainment staff left after calibration was introduced two years ago. The remainder reported relief at being judged on a score rather than background. - **The Algorithm's five steps (Tesla)** — Question every requirement. Delete every possible step. Simplify and optimise. Accelerate cycle time. Automate last. Cited from The Algorithm, a book by a former Tesla president, with the example of cutting the online car purchase from 64 steps to 13 by finding a bank willing to do one-page financing. - **Apple's seven engineering tiers** — Apple publishes seven engineer levels, each with a salary band; Bet-David said tier seven averages about 1.5 million dollars. The point: published levels eliminate conflict because everyone knows what the next rung requires and pays. - **Your money, my money, our money** — Bet-David's household agreement: each spouse has personal money for gifts and their own family's requests, and joint money for children and shared decisions. Nobody needs permission for their own money. Result: no financial arguments in the marriage. Used on stage as a model of managing expectations in advance. ## Exercises & frameworks to run - **Monday activity board** _(20 minutes every Monday)_ _[speaker]_ 1. Choose the two or three activities that actually produce results in each sales role, such as dials, appointments booked and appointments run. 2. Every Monday morning, have each person state last week's numbers for those activities out loud before mentioning any result. 3. Ask to see their calendar. Multiply appointments by average duration to estimate real hours worked. 4. Where the story and the calendar disagree, coach the activity gap first. Where they agree and results still lag, coach skill. - **Finish with:** A weekly activity scorecard per person and a habit of leading with inputs rather than excuses. - **Write your EALIR rubric** _(90 minutes, once)_ _[editorial (speaker idea, steps written by us)]_ 1. Take the five headings: effort, attitude, leadership, innovation, results. 2. Under each, write four observable criteria. Start from the speaker's list: effort = urgency, quality control, personal development, raising standards; attitude = perseverance, buy-in, culture fit, personal character; leadership = teamwork, developing others, confronting respectfully, accountability; innovation = creative and critical thinking, improvement, implementation; results = one-club reliability, hitting KPIs, beating your prior best. 3. For each criterion, write one sentence describing what a 3, a 7 and a 10 look like in your business. 4. Share the finished rubric with every manager before the first scoring round. - **Finish with:** A one-page scoring guide that lets two managers score the same person within a few points of each other. - **Three-view quarterly calibration** _(Three weeks per quarter, mostly managers' time)_ _[speaker]_ 1. Within a week of quarter close, every employee scores themselves 0-10 on each of the five categories with a specific example per score. 2. Each manager scores their direct reports independently on the same scale, without seeing the self-score first if you want a clean read. 3. The leadership committee meets. Each manager presents their scores. Leave 'meets' alone unless challenged. Debate every 'exceeds' and every 'does not meet' by asking 'why did you give this person that score?' 4. Agree a final number (for example self 47, manager 39, committee lands at 41). 5. The manager delivers the final score and the reasoning to the employee, opening with their self-score and the gap. - **Finish with:** A calibrated score out of 50 for every employee, a band (does not meet / needs improvement / meets / exceeds / outstanding), and a documented gap between self-view and company view. - **Manager distribution audit** _(30 minutes per quarter)_ _[editorial (speaker idea, steps written by us)]_ 1. After scores are collected but before the committee meets, tabulate each manager's percentage of reports in each of the five bands, plus the manager's own self-score. 2. Flag any manager with zero people below 'meets' as a possible people-pleaser. 3. Flag any manager whose self-score sits far above their team average as a possible elitist. 4. Compare each against the steward benchmark of roughly 10 / 20 / 50 / 15 / 5. 5. Bring the flagged distributions into the committee meeting and ask the manager to walk through their reasoning person by person. - **Finish with:** A one-page chart of every manager's rating shape and a short list of managers who need coaching or a skip-level check. - **The 47 versus 31 conversation** _(30 minutes per employee)_ _[speaker]_ 1. Open with the employee's own number: 'You scored yourself a 47. Tell me why you think you were a 47 last quarter.' 2. Listen fully. Then go to the numbers: which KPIs were hit, which were missed, what the calendar shows. 3. State your view plainly: 'At best I have you at 31.' Explain each category where the gap is widest. 4. Agree what a 40 would look like next quarter in specific, observable terms. 5. Write the agreed targets down and revisit them at the next calibration. - **Finish with:** A shared understanding of where the person stands, a closed expectation gap, and a written target for the next quarter. - **Skip-level talent check** _(One hour per month)_ _[editorial (speaker idea, steps written by us)]_ 1. Pick one team per month, prioritising teams whose manager's scores looked skewed or whose turnover is high. 2. Meet two or three of that manager's reports for 15 minutes each, without the manager present. 3. Ask: What are you working on? What gets in your way? How do you find out how you are doing? Who helps you grow? 4. Do not promise outcomes. Write down where what you hear differs from the manager's scores and story. 5. Feed the difference into the next calibration committee. - **Finish with:** A second-source read on team health and a list of potential future stars the manager may be under-rating. - **Publish your ladder** _(Half a day, once, then an annual review)_ _[editorial (speaker idea, steps written by us)]_ 1. List the levels that exist in your company for each major role family, the way Apple publishes seven engineering tiers. 2. For each level, write what the person must reliably do and the pay band attached. 3. Attach the calibration bands to money: which bands earn a raise, which earn a bonus multiplier, which earn nothing. 4. Share the document with the whole company and point to it whenever a raise or promotion is discussed. - **Finish with:** A public career and pay ladder that removes most arguments about fairness before they start. - **Delete steps before you automate** _(Two hours per process)_ _[editorial (speaker idea, steps written by us)]_ 1. Pick one process people complain about, such as your review cycle or your customer checkout. 2. Count every step a person has to take today. Tesla counted 64 to buy a car. 3. Run the five questions in order: Is this requirement real? Can this step be deleted? Can what remains be simplified? Can the cycle be shortened? Only then, what can be automated? 4. Set a target step count and redesign toward it. Tesla reached 13. - **Finish with:** A shorter process with a before-and-after step count you can show the team. ## What to do — and how often - [ ] **Hold a Monday activity review where each rep reports last week's calls and appointments before results, then checks the calendar.** — Monday morning, every week ('Go on the board Monday morning and tell me how many calls you made last week') _[weekly; speaker-stated]_ - [ ] **Write your five-category EALIR rubric with four criteria under each heading and share it with every manager.** — Once, before your first calibration cycle _[once; editorial recommendation]_ - [ ] **Run a full calibration: employee self-score, manager score, then a committee meeting that challenges every exceeds and does-not-meet.** — Every quarter ('Once a quarter managers come, we go through all of your employees') _[quarterly; speaker-stated]_ - [ ] **Have every employee score themselves on the five categories before their manager sees anything.** — Every quarter, as the first step of calibration _[quarterly; speaker-stated]_ - [ ] **Chart each manager's rating distribution against the 10/20/50/15/5 steward benchmark and flag people-pleasers and elitists.** — Every quarter, after scores are collected _[quarterly; editorial recommendation]_ - [ ] **Hold skip-level conversations with two or three reports of one manager whose scores look skewed.** — One team per month _[monthly; editorial recommendation]_ - [ ] **Give every employee scoring 11-24 (needs improvement) a written improvement plan and one quarter to move up.** — Each quarter someone lands in the band ('you have one quarter to clean up') _[quarterly; speaker-stated]_ - [ ] **Decide raises and bonus multipliers from the year's calibration bands: nothing for needs improvement, target for meets, an accelerator for exceeds and outstanding.** — Once a year, at year end ('if you score meets improvement by the end of the year, we don't pay you bonus') _[annually; speaker-stated]_ - [ ] **Publish a role ladder with levels, expectations and pay bands so nobody has to guess what the next step pays.** — Once this year, then reviewed annually _[once; editorial recommendation]_ - [ ] **Pick one bloated process and run the five-step algorithm (question, delete, simplify, accelerate, automate last) to cut its step count.** — Once per quarter, one process at a time _[quarterly; editorial recommendation]_ - [ ] **After each calibration, decide for every underperforming cluster whether the leak is the employee (coach) or the manager (invest or move out).** — Every quarter, at the end of the committee meeting _[quarterly; editorial recommendation]_ ## Questions to ask yourself - Is anyone on my team earning what I earn, or have I built a system where only I win? - When someone says 'I work hard', could I open their calendar and prove or disprove it in two minutes? - Who in my company is in the one club, and do they know how much I value them? - If every manager scored their team today, which of them would hand out zero low scores, and what am I going to do about that? - Which manager rates themselves far above their team, and how many future stars report to them? - Where would my employees score themselves versus where I have them, and how wide is the gap? - Do my people know exactly what score earns a raise, a multiplier or an exit, or do they find out after the fact? - Is the underperformance in my company clustered under one manager, and if so is the leak the employee or the manager? - Which recurring argument in my business or my home could a written agreement remove for good? ## Quotes > “I realize in that moment, everything's about systems, transfer of knowledge and actual accountability with numbers not tied to personality.” — Patrick Bet-David (17:02:37 · OycAYqFAC7qNpvLi2DKB) > > The moment he realised nobody on his sales team was earning six figures after seven years of personality-led management. > “Don't give me the final result. Start with your activity.” — Patrick Bet-David (17:03:48 · OycAYqFAC7qNpvLi2DKB) > > How he opened the Monday morning accountability board. > “We only have to ask you one time it gets done.” — Patrick Bet-David (17:09:25 · OycAYqFAC7qNpvLi2DKB) > > Defining the 'one club', the reliability marker under the results category. > “The more you're part of a one club, the more you're going to move up in the company.” — Patrick Bet-David (17:09:25 · OycAYqFAC7qNpvLi2DKB) > > Why reliability is the fastest path to promotion. > “We don't work with does not meet expectations, we just wish you nothing but the best.” — Patrick Bet-David (17:10:47 · OycAYqFAC7qNpvLi2DKB) > > The consequence of a 5-10 calibration score. > “It's black and white if you're a needs improvement.” — Patrick Bet-David (17:11:16 · OycAYqFAC7qNpvLi2DKB) > > No bonus and no raise for anyone ending the year in the 11-24 band, even if they threaten to leave. > “No, it's not realistic. But they're afraid of conflict.” — Patrick Bet-David (17:14:25 · OycAYqFAC7qNpvLi2DKB) > > On the people-pleaser manager who rates half the team outstanding and nobody below meets. > “So everybody self-calibrates themselves. And we say, really, you just scored yourself a 47. Tell me why you think you were 47 last quarter?” — Patrick Bet-David (17:16:44 · OycAYqFAC7qNpvLi2DKB) > > How the review conversation opens when the self-score is far above the manager's score. > “The elitist scores himself high. Everybody gets a horrible score.” — Patrick Bet-David (17:17:45 · OycAYqFAC7qNpvLi2DKB) > > The second manager type, whose teams lose future stars. > “So the elitist management gets exposed very early on, by the way. It's very easy. Actually, their first calibration they get exposed.” — Patrick Bet-David (17:19:22 · OycAYqFAC7qNpvLi2DKB) > > Why skip-level meetings plus an uncoached first calibration reveal bad managers quickly. > “Cuz now we're making decisions based on data and now we realize is the leak the employee or is the leak the middle management?” — Patrick Bet-David (17:21:18 · OycAYqFAC7qNpvLi2DKB) > > The diagnostic question calibration data lets a company answer. > “But if you don't have this data, how do you improve efficiency in your company? You don't.” — Patrick Bet-David (17:21:36 · OycAYqFAC7qNpvLi2DKB) > > Closing the case for measuring everyone. > “We exposed which managers we wanted to bet on and which ones we had to move out. And you know whose responsibility that was? Ours.” — Tom Ellsworth (17:22:46 · OycAYqFAC7qNpvLi2DKB) > > What two years of calibration data revealed about middle management at Valuetainment. > “Question every requirement, delete every possible step, simplify and optimize, accelerate cycle time, automate last.” — Patrick Bet-David (17:24:06 · OycAYqFAC7qNpvLi2DKB) > > Reciting the five-step algorithm from the book about Tesla, used as a model for cutting waste in any process. > “I manage expectations with everybody in my life, and that includes employees.” — Patrick Bet-David (17:28:51 · OycAYqFAC7qNpvLi2DKB) > > Tying the marriage money agreement, sponsor rules and calibration into a single habit. ## Watch-outs - Managing on personality and charisma for years while nobody on the team grows their income. - Accepting 'I work hard' or long hours in the office as evidence; only activity counts and the calendar tells the truth. - Paying salary, benefits and providing the leads without building any accountability mechanism to replace the market discipline of commission. - Rating everyone high to avoid conflict; it signals to the best people that performance does not matter and slows deadlines. - Letting an elitist manager who scores themselves high and everyone else low keep influence over nine reports until the stars quit. - Skipping the employee self-score, which leaves a hidden 15-point expectation gap that turns into resentment or a resignation. - Negotiating with a needs-improvement employee who threatens to leave over a missed raise; the bands only work if they are black and white. - Coaching managers on how to calibrate before their first round, which hides the biases you most need to see. - Treating calibration as an HR task; deciding which managers to bet on and which to move out is the founder's job. --- Cole's Notes Vault 2026 — https://vault.chels.ai — built by chels.ai (https://chels.ai). Only stage content is published; quotes are verbatim from automatic transcripts with light punctuation. Speaker-stated cadence and editorial recommendations are labelled separately. Times are Eastern; Las Vegas was three hours earlier.