Scoring goals fairly is harder than it sounds, mostly because "fair" gets defined differently by every manager doing the scoring unless there's a shared rubric. One manager scores generously because they don't want an awkward conversation. Another scores harshly because they think it motivates people. Neither is malicious, but the inconsistency is real, and employees notice it fast, especially when they compare notes with a peer on a different team.
This inconsistency corrodes trust in the whole goal-setting process faster than almost anything else. An employee who suspects their score depends more on which manager they report to than on their actual work has every reason to stop taking the exercise seriously, and that skepticism tends to spread across a team quickly once it takes hold.
Why a Shared Scale Matters More Than the Scale Itself
The specific numbers matter less than everyone using the same ones the same way. Google's widely used approach scores goals on a 0.0 to 1.0 scale, with an average of 0.6 to 0.7 treated as a healthy, ambitious result - not a failure. A team using a 1-10 scale, a percentage, or a simple red/amber/green system can work just as well, provided every manager applies it the same way.
A Simple, Practical Rubric
| Score | Meaning | What it should look like in practice |
|---|
| 0.0-0.3 | Little to no progress | Goal was barely started, or blocked without escalation |
| 0.4-0.6 | Partial progress | Real effort and movement, but the target wasn't reached |
| 0.7-0.9 | Strong result | Goal substantially achieved; this is the target zone for ambitious goals |
| 1.0 | Fully achieved | Consider whether the goal was ambitious enough if this happens often |
Bias Traps to Watch For
The relationship discount
Managers unconsciously score employees they like, or have worked with longer, more generously than employees they know less well - not out of malice, but because familiarity makes it easier to see effort and context that a shorter working relationship doesn't reveal. A rubric with specific, observable criteria reduces (though doesn't eliminate) this drift.
Recency bias
A strong final two weeks of the quarter can inflate a score that should reflect the whole period, just as a rough final stretch can unfairly drag down a quarter that was otherwise solid. Scoring against documented check-in notes from throughout the cycle, not just memory of the last few weeks, corrects for this.
The difficulty gap
Two employees hitting the same 0.8 score doesn't mean they did equally impressive work if one set a conservative, easily achievable goal and the other set a genuinely stretching one. Scoring in isolation, without discussing goal difficulty as part of the conversation, produces numbers that look fair but aren't.
Self-score anchoring
When employees self-score first, managers often anchor close to that number even when their independent assessment would differ, simply because it's socially easier to adjust slightly than to contradict outright. Having managers form an independent view before seeing the self-score - even briefly - produces more honest calibration.
Keeping Scoring Consistent Across Managers
The single most effective fix for cross-manager inconsistency is a short calibration session before scores are finalised - twenty minutes where managers briefly compare how they scored similar situations, before anything is locked in or shared with employees.
This surfaces the manager who's scoring everyone a full point higher than everyone else long before it becomes a company-wide fairness complaint.
It's also worth writing down, even briefly, why a score landed where it did - not a formal essay, a sentence or two. "0.6 - hit the launch date but adoption target fell short due to a delayed integration outside their control" gives future-you, or a different manager reviewing the history later, something to work from instead of a bare number.
Should Scores Be Tied to Compensation?
This is one of the most contested questions in goal-setting practice, and it's worth being direct about it: most OKR-native guidance recommends keeping goal scores separate from compensation and promotion decisions, specifically because tying them together pushes employees toward safe, easily achievable goals instead of ambitious ones. If your company does want goal outcomes to inform reviews, treat the score as one input among several - feedback received, qualitative context, goal difficulty - rather than a direct formula.
Scoring Goals That Aren't Naturally Numeric
Not every goal lends itself to a clean percentage. "Improve cross-team collaboration on the Lagos launch" is a real and important goal, but it doesn't have an obvious number attached to it the way "grow monthly active users to 50,000" does.
For goals like this, it helps to define, at the time the goal is set, two or three concrete, observable signs of success - specific enough that two different managers reviewing the same evidence would likely land on a similar score. Waiting until the end of the cycle to figure out what "good" would have looked like almost guarantees inconsistent, retroactively-justified scoring.
A practical fix many teams use is converting qualitative goals into a short checklist scored at the end: did the launch plan get reviewed by both teams before execution, did the post-launch retro happen, did the other team's lead confirm collaboration felt smoother than the previous cycle. None of these are perfectly objective, but they're far more consistent than scoring "how collaborative did this feel" from memory months later.
How Often Rubrics Should Be Revisited
A scoring rubric isn't meant to be permanent. As a company matures its goal-setting practice, it's worth revisiting the rubric every few cycles - not to change it constantly, which reintroduces the inconsistency problem, but to check whether the scale is still producing a healthy spread of scores rather than clustering unrealistically high or low. If every manager's average score has crept up toward 0.9 or 1.0 over several cycles without a corresponding jump in actual company performance, that's usually a sign scoring has drifted generous over time, and a recalibration conversation is overdue.
Where Talstack Fits
Talstack's Goals module supports whichever scoring scale a team chooses, and keeps a running history of check-in notes attached to each goal, not just a final number - which makes the "why did this land here" context available months later, not just in the scoring manager's memory. Because Goals sits alongside 360 Feedback in the same Talstack platform, a manager reviewing a score can see the feedback an employee received across the cycle as additional context, rather than scoring a goal in isolation from everything else that happened during it.
Handling Disagreement Between Manager and Employee
Occasionally a manager's score and an employee's own view of their performance genuinely diverge, even after a good-faith conversation. When this happens, it's worth separating two different questions: was the process fair (did the employee have clear expectations, regular feedback, and a genuine chance to succeed), and was the outcome fair (does the score accurately reflect what happened). An employee can accept a lower score more easily when they trust the process was fair, even if they'd have scored themselves slightly higher - most disputes are really about a breakdown in one of these two things, not a fundamental disagreement about the facts.
Giving employees a structured way to note disagreement - a short comment attached to their score, rather than an informal complaint that goes nowhere - also matters. It doesn't have to change the outcome, but having a documented voice in the record tends to reduce the sense that scoring is something done to them rather than with them.
Training New Managers to Score Consistently
A newly promoted manager scoring goals for the first time is one of the most common sources of scoring drift, simply because they haven't yet been through a calibration cycle or absorbed the informal norms more experienced managers have picked up over time.
Pairing a first-time scorer with a more experienced manager for their first cycle - not to override their judgment, but to talk through a few scores together before they're finalised - closes this gap faster than a written policy document alone ever will. Most scoring inconsistency isn't due to bad faith; it's due to managers genuinely not knowing what "normal" looks like yet.
Quick Checklist: Scoring Goals Fairly
- Agree on one scoring scale across the whole company
- Hold a short calibration session before finalising scores
- Score against documented check-in notes, not end-of-cycle memory
- Form an independent view before seeing the employee's self-score
- Write a one-line reason alongside every score
- Decide explicitly whether scores feed compensation, and communicate that decision clearly
Copy-Paste Script: Introducing a Scoring Rubric to Managers
Subject: Standardising how we score goals this cycleHi team,Before we finalise scores this cycle, I want us all using the same rubric, so an employee's score reflects their work, not which manager happened to review it.The scale:- 0.0-0.3: little to no progress- 0.4-0.6: real, partial progress- 0.7-0.9: strong, substantial achievement (this is where most solid results should land)- 1.0: fully achieved - if this is common across your team, our goals may not be ambitious enoughBefore you finalise anyone's score, jot a one-line reason for it, and let's do a quick 20-minute calibration call as a management team before anything goes final.Thanks,[HR/People lead name]
Frequently Asked Questions
What's a fair scoring scale for goals?
There's no single correct scale - a 0.0-1.0 scale, a percentage, or a simple tiered system can all work - what matters most is that every manager applies the same scale the same way, which usually requires a short calibration process.
Is it bad if most people score 1.0 on their goals?
Usually, yes. Consistently perfect scores across a team often signal that goals weren't ambitious enough rather than that performance was flawless - a healthy average tends to sit lower, around 0.6-0.7 on a 1.0 scale.
Should employees self-score before their manager scores them?
It's fine to include self-scoring as part of the process, but managers should form an independent view first to avoid simply anchoring to the employee's number, which tends to compress genuine disagreement out of the conversation.
How do you keep scoring consistent when different managers score different teams?
A brief calibration session - managers comparing how they scored similar situations before finalising anything - is the most effective practical fix for cross-manager inconsistency.
Should a missed goal always result in a low score?
Not necessarily - a goal missed due to factors outside the employee's control (a delayed dependency, a shifted business priority) deserves a different score and a different conversation than a goal missed due to lack of effort.
Should goal scores directly determine bonuses or raises?
Most goal-setting frameworks recommend against a direct formula, since it incentivises conservative goal-setting. Treating the score as one input alongside feedback and context tends to produce fairer, more honest outcomes.
How should a new manager learn to score goals consistently?
Pairing a first-time scorer with a more experienced manager for their first cycle, to talk through a few scores together before finalising, closes the calibration gap faster than a written policy alone.
