Lesson 9 of 16 · Level 3: Advanced
Goals and OKRs for Your Team
- ⏱ 12 min read
- ❓ 4 review questions
- ✍️ Exercise with solution
- 📥 Downloadable template
In this lesson
- How to write SMART goals and where the acronym came from.
- What OKRs (Objectives and Key Results) are and how Intel and Google used them.
- How to tell a strong key result from a to-do item.
- How to calculate key result progress, including metrics that should go down, and roll it up into one objective score.
- How to run OKRs week to week without turning them into paperwork.
Enter each key result's start, target and current value: the sheet calculates progress, weighted objective progress, the share of the quarter elapsed and an on-track status.
- leadership-toolkit.xlsx · sheet OKR Tracker
Clear goals are the cheapest performance tool a leader has: people can't hit a target they can't see. OKRs (Objectives and Key Results) are a goal-setting framework that pairs an inspiring, qualitative objective with a few measurable results that prove you got there. In this lesson you'll write OKRs for a real team, learn the formula for tracking their progress and work through a full calculation from a Chicago warehouse.
Start with SMART goals
The SMART acronym first appeared in a 1981 Management Review article by George T. Doran, "There's a S.M.A.R.T. Way to Write Management's Goals and Objectives." His original letters stood for Specific, Measurable, Assignable, Realistic and Time-related. Today you'll often see "Achievable" and "Relevant" instead, but the idea is the same: a goal should be clear enough that two people would agree on whether it was met.
| Vague goal | SMART version |
|---|---|
| "Improve customer service." | "Raise our average post-call survey score from 4.1 to 4.4 by December 31, with Ana owning the coaching plan." |
| "Reduce waste in the kitchen." | "Cut weekly food waste from 180 lb to 140 lb by the end of Q2, measured on the Friday weigh-in." |
What OKRs are
OKRs grew out of Andy Grove's management system at Intel in the 1970s. John Doerr learned them there and introduced them to Google's founders in 1999, then described the method in his book Measure What Matters (2018). Doerr's one-line formula: "I will (Objective) as measured by (this set of Key Results)."
- Objective: what you want to achieve. Qualitative, memorable, motivating. "Make our warehouse the most reliable link in our customers' supply chain."
- Key results: how you'll know you got there. Numeric, with a starting value, a target and a deadline. "Raise on-time shipments from 91% to 97%."
- Initiatives: the projects and tasks you'll do to move the key results. They're not key results themselves.
Google's re:Work guide recommends picking just three to five objectives, since more can spread a team too thin, with around three key results each. It also describes OKRs as ambitious by design, graded from 0.0 to 1.0, with an expected average of about 0.6 to 0.7. If you consistently score 1.0, the goals probably aren't ambitious enough. Some organizations separate committed OKRs (must hit 1.0, like a contractual service level) from aspirational ones (stretch goals where 0.7 is a good result). Tell your team which is which.
- Pick one problem that matters
- Write the objective in plain words
- Add 3–4 measurable outcomes
- Record start and target values
- List initiatives separately
1. Which of these is a well-written key result?
It's measurable, has a start and a target value, and describes an outcome. Launching a program is an initiative (a task), and "be the best warehouse" is closer to an objective.
How to calculate key result progress
Progress measures how much of the distance between the starting value and the target you've covered. One formula works whether the metric should go up or down:
Progress = (Current − Start) ÷ (Target − Start)
Capped between 0% and 100%
For a metric that should go down, both the top and the bottom of the fraction are negative, so the result is still positive. Then roll the key results up into one objective score, either with a simple average or with weights that reflect importance:
Objective progress (weighted) = Σ (KR progress × KR weight) ÷ Σ weights
Worked example: a Chicago warehouse, Q3
Dana Kowalski's team set this objective for the third quarter (July 1 to September 30): "Make our Chicago warehouse the most reliable link in our customers' supply chain." Here's the status on August 26:
| Key result | Start | Target | Current | Calculation | Progress | Weight |
|---|---|---|---|---|---|---|
| KR1: On-time shipment rate | 91% | 97% | 95.2% | (95.2 − 91) ÷ (97 − 91) = 4.2 ÷ 6 | 70% | 40% |
| KR2: Picking error rate (lower is better) | 0.80% | 0.30% | 0.50% | (0.50 − 0.80) ÷ (0.30 − 0.80) = −0.30 ÷ −0.50 | 60% | 20% |
| KR3: Dock-to-stock time (lower is better) | 48 h | 24 h | 30 h | (30 − 48) ÷ (24 − 48) = −18 ÷ −24 | 75% | 20% |
| KR4: Associates cross-trained on 2 stations | 0 | 12 | 6 | (6 − 0) ÷ (12 − 0) | 50% | 20% |
Simple average: (70% + 60% + 75% + 50%) ÷ 4 = 63.75%
Weighted: 70%×0.4 + 60%×0.2 + 75%×0.2 + 50%×0.2
= 28% + 12% + 15% + 10% = 65%
Quarter elapsed: 56 days ÷ 91 days = 61.5%
Comparing progress with time elapsed tells Dana where to focus. The toolkit uses a simple rule: on track if progress is at or above the share of the quarter elapsed, at risk if it's up to 15 points below and off track if it's further behind. KR1 (70%) and KR3 (75%) are on track. KR2 (60%) and KR4 (50%) are at risk, since both fall between 46.5% and 61.5%. Her next team meeting focuses on two questions: what's blocking error reduction, and how to free associates for cross-training.
In Excel or Google Sheets (start in D, target in E, current in F, weight in G):
Progress (H5): =IF(E5=D5,"",MAX(0,MIN(1,(F5-D5)/(E5-D5))))
Objective: =SUMPRODUCT(H5:H8,G5:G8)/SUM(G5:G8)
Elapsed: =MAX(0,MIN(1,(AsOfDate-StartDate)/(EndDate-StartDate)))
2. A call center KR is "Reduce average hold time from 6 minutes to 2 minutes." Hold time is now 3.5 minutes. What's the progress?
(3.5 − 6) ÷ (2 − 6) = −2.5 ÷ −4 = 0.625, or 62.5%. The team has covered 2.5 of the 4 minutes it set out to cut. 37.5% is the distance still left.
3. An objective has three KRs at 80%, 50% and 20% progress, weighted 50%, 30% and 20%. What's the weighted objective progress?
80% × 0.5 + 50% × 0.3 + 20% × 0.2 = 40% + 15% + 4% = 59%. The simple average would be 50%; weighting gives more credit to the most important KR.
Running OKRs through the quarter
- Week 0Set and shareDraft with the team, not for the team. Publish the final OKRs where everyone can see them, with an owner for each key result.
- Every week · 15 minUpdate and talkEach owner updates the current value and answers: "What moved, what's stuck, what do you need?"
- Mid-quarterReview and adjustCompare progress with time elapsed. Drop or rewrite a KR only if the world changed, not because it's hard.
- Quarter endScore and reflectGrade each KR from 0.0 to 1.0, then ask what you learned and what carries over.
A script for the weekly check-in
- "KR2 is at 60% and we're 62% through the quarter. On a scale of 1 to 10, how confident are you we'll hit 0.30% by September 30?"
- "What's the one thing that would move this most in the next two weeks?"
- "What should we stop doing to make room for it?"
Google's re:Work guide states that OKRs are not synonymous with employee evaluations, and What Matters, the organization founded by John Doerr, advises that OKRs should not factor into compensation or bonuses. When pay depends on hitting a number, people set easy targets ("sandbagging"), and the ambition that makes OKRs useful disappears.
Ten objectives "because everything is important."
Three to five at most for a team. If everything is a priority, nothing is.
Key results that are tasks: "Hold four trainings."
Measure the outcome the trainings should cause: "Cut errors from 0.80% to 0.30%."
Setting OKRs and looking at them again at quarter end.
Update weekly; a 15-minute check-in keeps them alive.
Treating a 0.7 on an aspirational OKR as failure.
Celebrate it. A consistent 1.0 means the bar was too low.
4. A team scores 1.0 on every aspirational OKR for three quarters in a row. What does that most likely mean?
Aspirational OKRs are designed so that a full score is rare. Google's guidance expects an average around 0.6 to 0.7; a steady 1.0 suggests the team is sandbagging or playing it safe.
Practice exercise
Priya's remote team in Seattle has this objective: "Make our mobile checkout feel instant." It's the end of week 6 of a 13-week quarter. Calculate each KR's progress (equal weights), the objective progress and each KR's status using the on-track rule from this lesson.
| Key result | Start | Target | Current |
|---|---|---|---|
| KR1: Checkout load time (p95, seconds) | 3.2 | 1.5 | 2.1 |
| KR2: Crash-free sessions | 98.0% | 99.5% | 99.2% |
| KR3: App store rating | 4.1 | 4.6 | 4.2 |
See solution
- KR1: (2.1 − 3.2) ÷ (1.5 − 3.2) = −1.1 ÷ −1.7 = 64.7%
- KR2: (99.2 − 98.0) ÷ (99.5 − 98.0) = 1.2 ÷ 1.5 = 80%
- KR3: (4.2 − 4.1) ÷ (4.6 − 4.1) = 0.1 ÷ 0.5 = 20%
Objective progress: (64.7% + 80% + 20%) ÷ 3 = 54.9%.
Time elapsed: 6 ÷ 13 = 46.2%. The at-risk threshold is 46.2% − 15 points = 31.2%.
Status: KR1 on track (64.7% ≥ 46.2%), KR2 on track (80%), KR3 off track (20% is below 31.2%). App ratings lag behind performance fixes, so Priya might add an initiative such as an in-app prompt asking satisfied users to leave a review, and check whether the rating reflects issues the team isn't measuring.
Summary
- SMART comes from George T. Doran (1981): specific, measurable, assignable, realistic, time-related.
- OKRs pair a qualitative objective with around three measurable key results; initiatives are separate.
- Progress = (current − start) ÷ (target − start), capped at 0–100%, for metrics that go up or down.
- Roll up with a simple or weighted average, and compare with the share of the quarter elapsed.
- For aspirational OKRs, about 0.6–0.7 is a good average. Keep OKRs out of pay decisions.
Next lesson: Difficult Conversations, Step by Step. You'll prepare and run the conversations most managers put off, with scripts for the toughest moments.
Create your account in 30 seconds, mark your lessons and download your certificate at the end.