The central problem in organizational management is the gap between strategy and execution. Every company has a strategic direction — a theory of where it’s going and why. Most of them fail to connect that strategic direction to the day-to-day work of the people building toward it. The result is organizations where the executive team is aligned on direction and the operating teams are pulling in different or merely adjacent directions because no one translated the strategy into the specific things each team should actually be doing.
What OKRs Actually Are
OKRs are not complicated. The framework has two components:
Objectives: What you want to achieve. They should be significant, concrete, action-oriented, and inspirational. Good objectives create direction and meaning. They answer the question: what are we trying to accomplish this period?
Key Results:
How you know you’re achieving the objective. They should be specific, time-bound, aggressive but realistic, and above all measurable. Key results are not tasks or activities — they’re outcomes. The distinction is critical. “Launch the new product” is a task. “Achieve 10,000 downloads by December 31” is a key result. The first tells you what to do. The second tells you whether what you did worked.
Together, they create a feedback loop: the objective gives direction, the key results tell you whether you’re moving in that direction, and the combination forces clarity about what success actually looks like.
Doerr learned the system from Andy Grove, who developed it at Intel as a refinement of Peter Drucker’s Management by Objectives framework. Grove’s insight was that the measurement component — the key results — was what made the difference. Drucker’s MBO often produced goal-setting without accountability, because the goals were frequently vague enough to declare success regardless of outcomes. Grove insisted on specific, measurable outcomes and created the discipline of reviewing progress against them at regular intervals.
“Ideas are easy. Execution is everything. It takes a team to win. OKRs have helped lead us to ten-fold growth, many times over.”
The Four Superpowers of OKRs
Doerr organizes the benefits of OKRs around what he calls four “superpowers.” Not mystical properties — specific ways the system changes organizational behavior:
- Focus and Commit to Priorities. The OKR process forces organizations to choose. You can’t have twenty-five objectives. The discipline of identifying three to five objectives per quarter — and committing resources to them — forces the prioritization that most organizations avoid. This is where the system’s friction becomes valuable: if everything is a priority, nothing is.
- Align and Connect for Teamwork. OKRs are typically transparent across the organization. When individuals and teams can see each other’s objectives and key results, coordination improves and misalignment becomes visible. The sales team can see what the product team is working toward. The engineering team can see what the customer success team needs. Dependencies and conflicts surface earlier, when they’re still manageable.
- Track for Accountability. The measurement component creates accountability that vague goals don’t. When your key result is “achieve $2M in pipeline by Q3,” you know at any point in the quarter whether you’re on track. You can’t claim success on a fuzzy objective. You can recalibrate mid-quarter if circumstances change. The regular check-in cadence — weekly, monthly, quarterly — keeps the system alive rather than letting it become an annual ritual of goal-setting and forgetting.
- Stretch for Amazing. Doerr and Grove both argue for “stretch goals” — objectives you might not hit, set at a level that requires exceptional performance. The conventional wisdom is that you should set goals you can achieve. The OKR philosophy is that goals you’re certain you’ll achieve are probably not ambitious enough. Google targets a 0.7 score on its key results as success — 70% of a stretch goal is more valuable than 100% of a comfortable one.
The Intel Origin Story
The most valuable sections of Measure What Matters are the historical ones — specifically the account of Andy Grove implementing OKRs at Intel in the 1970s and the effect on one of the most significant competitive battles in technology history.
In 1980, Motorola’s 68000 microprocessor was widely considered superior to Intel’s 8086. IBM was choosing a chip for the IBM PC — a product that would define the personal computer market. Motorola was the obvious choice. Intel launched “Operation Crush,” a wartime-style campaign to win the IBM design win at any cost. The campaign used OKRs as its coordination mechanism — specific objectives (win the IBM design), measurable key results (number of design wins per month), aligned across sales, product, and marketing teams with weekly check-ins.
Intel won the IBM design. The 8086 architecture became the dominant personal computer standard. The rest is history. The OKR system wasn’t the reason Intel won — the company had excellent engineers, strong relationships, and legitimate technical arguments. But the OKR system created the organizational alignment that let those assets get deployed coherently toward a specific goal. Without that alignment, a diffuse competitive effort might have produced worse results even with the same underlying capabilities.
This is the most honest case for OKRs: not that the system produces results by itself, but that it prevents organizations from wasting the results-producing capabilities they already have. Alignment doesn’t generate capability — it ensures capability isn’t squandered.
Where OKRs Break Down
Doerr is more honest about failure modes than many OKR advocates, though not as honest as he could be. The main risks he identifies:
Too many OKRs:
When every department and team has fifteen objectives, the focus function disappears. The discipline of OKRs requires the discipline of saying no, which means organizational politics will constantly push against the system by trying to add objectives that are important to various stakeholders.
OKRs as performance reviews:
The most common implementation mistake. When OKR scores feed directly into compensation and promotion decisions, people stop setting stretch goals and start sandbagging. The stretch goal value disappears entirely. Doerr is explicit: OKRs should inform but not determine performance reviews. They’re a tool for organizational learning and coordination, not a scoring system for individual judgment.
Superficial implementation:
Setting OKRs once, never reviewing them, and declaring success at year-end is worse than not having OKRs at all. It creates the illusion of alignment while producing none. The system only produces value through the cadence — the regular check-ins that maintain connection between the goals and the work.
The failure mode Doerr doesn’t fully address is the one that actually kills most OKR implementations: senior leadership that treats OKRs as something for the operating teams rather than for themselves. When executives set vague objectives and specific teams are expected to produce measurable key results against them, the system functions as a measurement and accountability tool for lower levels of the organization while leaving strategic clarity unaddressed at the top.
The tool is being used to manage the symptom (execution drift) rather than the disease (strategic ambiguity).
The Google Case Studies
A significant portion of Measure What Matters consists of case studies — Google, Intel, Bono’s ONE campaign, the Gates Foundation, and several others. Well-constructed and genuinely informative, but they create a subtle problem: all the case studies are from exceptional organizations with abundant resources, strong leadership, and missions that inspire genuine commitment from employees. The OKR system at Google operated in a context where the people implementing it were highly intelligent, intrinsically motivated, and working on problems they found genuinely meaningful.
The transferability question — does OKRs work as well for a mid-sized manufacturing company or a regional bank or a nonprofit with high turnover — is never seriously addressed. The system probably still produces value in those contexts, but the specific mechanisms (transparent cross-team objectives, stretch goals, high-frequency check-ins) require significant adaptation for organizations that don’t look like Silicon Valley tech companies circa 2005.
CFRs: The Underemphasized Companion System
- Conversations: Regular, high-quality one-on-one exchanges between managers and employees about progress, obstacles, development, and context. Not performance reviews — ongoing conversations that keep the manager connected to what’s actually happening in the work.
- Feedback: Continuous, specific, bidirectional feedback rather than annual reviews. Both managers giving feedback to employees and employees giving feedback to managers. The bi-directionality matters — it’s what creates the information flow that makes the OKR data usable.
- Recognition: Systematic acknowledgment of contributions, including contributions that don’t show up in key results. Not just the person who hit their number — also the person who helped someone else hit theirs, who identified a problem before it became a crisis, who maintained culture quality under pressure.
Late in the book, Doerr introduces what he calls CFRs — Conversations, Feedback, Recognition — as the human management complement to OKRs. This section deserves more prominence than it gets.
The argument: OKRs are a goal-setting and coordination system, not a management system. They tell you and your team what you’re working toward and whether you’re getting there. They don’t tell you how to develop your people, how to have difficult performance conversations, how to recognize contributions that didn’t move a metric, or how to build the psychological safety that allows honest check-ins. CFRs are the management practices that give the OKR data meaning.
The CFR system is less flashy than OKRs but arguably more important for the kinds of organizational improvements most companies need. Implementing OKRs without CFRs produces measurement without development — you know what’s happening but you’re not building the capacity for it to go better.
“When you measure what matters, you improve what matters. The key is knowing what matters.”
Applying OKRs Without the Book’s Blind Spots
A practical implementation guide that addresses the gaps in Doerr’s treatment:
- Start with strategy, not with OKRs. OKRs are a translation tool — they translate strategy into measurable objectives. If your strategy is unclear, OKRs will efficiently align your team toward the wrong things. Fix the strategy first (see Rumelt). Then use OKRs to connect it to execution.
- Executives set OKRs first. The direction-setting function of OKRs only works if it flows from the top down before it flows from the bottom up. If the CEO doesn’t have three clear, measurable objectives for the quarter, team OKRs will be random acts of measurement rather than aligned action.
- Separate OKRs from compensation, completely. The stretch goal function dies immediately when OKR scores determine bonuses. Use OKRs for organizational learning. Use separate criteria for compensation. The two systems have different purposes and mixing them destroys the more important one.
- Build the check-in cadence before worrying about the objectives. The most common implementation failure is excellent goal-setting followed by no follow-through. Before you set your first OKR, define who will check on progress, how frequently, and what format the check-in will take. The cadence is the system. The goals are just its input.
For the broader context of goal-setting and performance, our piece on self-discipline as a foundation of resilience explores the psychological dimension of sustained performance. The connection between clear goals and mental focus is explored in our guide to focus and concentration. And for how high-performing teams use accountability structures, see our piece on building resilience in the workplace.
The OKR Implementation Checklist: What Most Guides Skip
Doerr’s book describes the OKR system comprehensively but spends less time on the specific organizational conditions determining whether an OKR implementation succeeds or fails. Experience with failed implementations reveals what those conditions actually are.
First, OKRs require strategic clarity to be useful. If the company’s strategic direction is ambiguous — if the executive team disagrees, implicitly or explicitly, about priorities — OKRs will efficiently align teams toward a priority that shifts the next time the strategic ambiguity surfaces. Implementing OKRs before resolving strategic ambiguity doesn’t create alignment. It accelerates the speed at which misalignment becomes visible, which is useful for diagnosing the problem but should not be mistaken for solving it.
Second, the cadence is the system. The objectives and key results are just inputs. The value comes from the weekly check-ins, the monthly progress conversations, the quarterly retrospectives asking “what did we learn from the gap between target and actual?” Without those review rituals, OKRs degrade from an alignment and learning system into an annual goal-setting exercise with fancier vocabulary. Most failed OKR implementations have excellent goals and absent cadence.
Third, leadership must model before mandating. If the CEO’s objectives are vague, if the executive team’s key results aren’t actually measurable, if senior leaders treat OKR review meetings as administrative obligations rather than substantive conversations — the operating teams will observe and calibrate their own engagement accordingly. The signal from the top that OKRs are how serious work gets aligned in this organization has to be genuine, not just declared. Declaration without the model is one of the most reliable ways to produce cynical compliance rather than genuine adoption.
Fourth, OKR scores and performance reviews must be formally and visibly separated. The moment people believe career advancement depends on their OKR score, the stretch goal function dies. Ambitious targets get replaced by sandbagged ones. Honest retrospective analysis gets replaced by score justification. The entire learning mechanism that makes OKRs valuable evaporates. This separation has to be not just stated but demonstrated — through visible examples of people who scored 0.5 on ambitious goals and received recognition for the attempt, and through visible examples of OKR scores not being cited in performance conversations.
OKRs in Practice: The Gates Foundation Case
One of the most instructive case studies in the book — and one receiving less attention than the Google examples — is the Gates Foundation’s adoption of OKRs. The Foundation’s context is instructively different from Google’s: the outputs are harder to measure (improving global health is not a quarterly metric), the time horizons are longer (vaccination campaigns take years to evaluate), and the “customers” have no ability to express preference through market behavior.
The adaptation the Foundation made to standard OKR practice is worth understanding. The key results in a traditional technology OKR context are typically leading indicators that predict the desired outcome: X new users, Y revenue, Z customer satisfaction score. For global health initiatives, the equivalent leading indicators are things like: number of countries with vaccination rates above threshold, coverage of target population with preventive treatment, reduction in disease incidence in target geography. These are harder to observe, take longer to materialize, and require more sophisticated measurement infrastructure to track reliably.
The Gates Foundation case demonstrates that the OKR framework is flexible enough to work in contexts very different from its Silicon Valley origins, but that doing so requires genuine intellectual investment in identifying the right key results rather than the measurable proxy that happens to be available. The difference between a key result that actually tracks progress toward the objective and one that merely counts an activity related to the objective is the difference between an OKR implementation that produces learning and one that produces measurement theater.
The Relationship Between OKRs and Intrinsic Motivation
Doerr discusses the motivational benefits of OKRs but doesn’t fully develop the most important insight about the relationship between goal-setting and motivation. The research on this topic — from self-determination theory to the goal-setting literature — suggests that goals have predictably different effects on motivation depending on how they relate to the person’s intrinsic values and sense of autonomy.
Goals experienced as externally imposed — assigned by a manager, driven by organizational pressure, misaligned with the person’s own priorities — produce compliant behavior but reduced intrinsic motivation. Goals experienced as self-determined — chosen in pursuit of something genuinely cared about — produce engagement and persistence externally-imposed goals can’t match. The OKR literature, including Doerr’s book, describes the goal-setting mechanics without fully addressing this distinction.
The practical implication: the OKR system works best when individual key results get determined primarily by the person responsible for them, with organizational objectives providing the context and direction but the specific metric shaped by individual judgment about what matters most and what can genuinely be influenced. When OKRs get handed down from the top with key results pre-specified, the system captures the alignment benefits while sacrificing the motivation benefits. The design that captures both requires genuine participation in key result definition, not consultation theater where the results are already decided before the conversation happens.
The Superpower of Alignment: Why It Matters More Than You Think
Doerr’s case for alignment is more profound than it initially appears. Most discussions of alignment treat it as a coordination mechanism — a way to ensure different parts of the organization are working toward compatible rather than conflicting goals. Correct but incomplete. The deeper value of alignment is what it does to the organization’s collective intelligence.
Consider an organization where every team is working hard but toward slightly different interpretations of what success looks like. Marketing optimizing for brand awareness because their version of the strategic priority is “expand recognition.” Sales optimizing for volume because their version is “grow revenue fast.” Product optimizing for feature breadth because their version is “be the most capable solution.” Customer success optimizing for retention because their version is “build long-term relationships.” Each team competent at its own job. The organization’s total output significantly less than the sum of its parts, because the parts are pointed in different directions.
OKR alignment doesn’t just reduce this coordination waste. It changes what information flows to the top and what decisions get made. When teams are aligned on specific measurable objectives, the failures and near-misses that reveal strategic errors surface faster. A team pursuing a clearly-defined key result that isn’t moving generates a visible signal — something is wrong with either the approach or the assumption underlying the objective — that a team pursuing a vague goal does not generate, because vague goals can always be described as in-progress.
The strategic implication: OKR alignment is not primarily a coordination tool. It’s a learning tool. The value is not just that teams stop working at cross-purposes — it’s that the specific, measurable objectives create a feedback loop that reveals strategic assumptions that are wrong, early enough to correct them before they’ve consumed a year of organizational effort.
Quarterly vs. Annual OKR Cycles: The Tempo Question
Doerr advocates strongly for quarterly OKR cycles rather than annual ones, and the reasoning deserves more development than the book gives it. The tempo of goal-setting cycles is not an administrative detail — it determines how quickly the organization learns from the gap between target and actual, which determines how quickly strategic errors are corrected.
Annual goal cycles were calibrated to the pace of organizational change in the mid-twentieth century, when strategy shifted slowly and the main challenge was executing a known plan against a stable competitive environment. In environments where competitive dynamics shift quarterly — as most technology and consumer markets do — annual cycles are too slow. An assumption that was wrong in January and produced poor results through March is already three months late to correction if the next strategic conversation happens in January of the following year.
The quarterly cycle is not universally correct either. Some objectives — long-term capability development, infrastructure investments, relationship building — operate on longer timescales where quarterly measurement of progress would either produce false urgency or require such fine-grained leading indicators that the measurement overhead exceeds its value. The correct answer is usually a tiered system: annual or multi-year objectives for the longest-horizon strategic priorities, quarterly OKRs for the operational priorities that need frequent feedback loops, and monthly or weekly metrics for the leading indicators that allow mid-quarter course correction. The OKR framework is flexible enough to accommodate this tiering; Doerr’s presentation of it as primarily a quarterly system is a simplification that some readers will need to adapt.
What OKRs Can’t Fix
The most useful question to ask about any management framework is: what does it not address? OKRs are a powerful tool for focus, alignment, accountability, and organizational learning. They don’t address — and shouldn’t be expected to address — several things that determine organizational performance at least as significantly.
They don’t fix culture. An organization with a culture of blame, fear, or political competition will implement OKRs as a political competition system. The objectives will be set to protect departmental turf. The key results will be negotiated to minimize accountability while maximizing credit. The retrospectives will be exercises in score justification rather than honest analysis. OKRs in a broken culture accelerate the expression of the dysfunction; they don’t address the dysfunction itself.
They don’t fix strategy. As noted earlier, OKRs require strategic clarity as an input. They can’t generate it. An organization using OKRs to generate the appearance of strategic direction — “our objective is to be the leading provider of X to Y customers” — is using the format to substitute for the underlying strategic thinking. The key results will then measure progress toward a poorly-specified objective, which is worse than no measurement at all, because it produces false confidence that the strategic question has been answered.
They don’t fix leadership. A manager who sets unclear expectations, provides inadequate feedback, fails to develop their people, and makes poor decisions will do all of these things within an OKR framework as surely as they do them without one. The framework changes the format of the dysfunction, not the substance of it. A team with poor leadership and OKRs has different problems than a team with poor leadership and no OKRs, but they both have the same underlying problem.
Understanding what OKRs can and cannot fix is the key to using them well. Apply them where their specific mechanisms — forcing prioritization, creating measurement, building review cadence, enabling transparency — address real organizational gaps. Don’t apply them as a substitute for the more difficult work of strategic clarity, culture development, and leadership quality that they require as preconditions.
The Long Shadow of Andy Grove
Any honest assessment of Measure What Matters has to acknowledge that the most important ideas in it are not Doerr’s. They’re Grove’s. Grove developed the OKR framework at Intel in the 1970s as a refinement of Drucker’s Management by Objectives. Doerr learned the system from Grove, deployed it at Kleiner Perkins and in companies like Google where he invested, and wrote this book thirty years after Grove’s original work.
Not a criticism of Doerr — he attributes the framework clearly to Grove throughout the book. But it means that anyone wanting to understand the underlying management philosophy that makes OKRs work should treat Grove’s High Output Management as a more complete and more rigorous source than this book. Doerr provides more modern case studies and a more accessible introduction. Grove provides the why at a level of depth Doerr doesn’t match.
The serious student of the topic reads both. Start with Measure What Matters for an accessible introduction to the framework and compelling modern examples. Return to High Output Management to understand the production-management philosophy that makes the framework make sense at a deeper level. Reading Grove after Doerr reveals dimensions of Grove’s thinking that weren’t visible before — the way the OKR system fits into a broader theory of management use, measurement, and organizational output that is Grove’s most enduring contribution to the field.
OKRs and Remote Work: The Renewed Relevance
One dimension of OKRs that Doerr couldn’t have emphasized in 2018 but that has become increasingly important is their particular value in distributed and remote work environments. The alignment and transparency functions of OKRs become more critical, not less, when teams are not physically co-located. In an office, alignment happens partly through osmosis — you overhear conversations, observe what people are working on, have informal interactions that surface misalignments before they become problems. In distributed environments, none of this happens. The misalignments that osmosis would have caught accumulate unnoticed until they’re large enough to be unmissable.
OKRs in distributed contexts provide the explicit coordination mechanism that proximity used to provide implicitly. When everyone’s objectives and key results are visible — not just to their immediate team but across the organization — people can see what others are working toward, identify potential interdependencies and conflicts before they materialize, and offer relevant help they wouldn’t have known to offer otherwise. The transparency function of OKRs is arguably more valuable in remote work than in any other context, because the alternative to explicit transparency is not reduced but rather invisible information asymmetry.
The implementation challenge in remote contexts is maintaining the check-in cadence without the physical cues that office environments provide. Without the visual reminder of the OKR board on the wall, without the hallway conversations that naturally surface progress and obstacles, the check-in cadence needs to be more deliberately structured and more consistently enforced. Asynchronous weekly check-ins (brief written updates on key result progress and current obstacles) combined with synchronous monthly OKR reviews have emerged as a workable format for many distributed teams — substantive enough to maintain the feedback loop, lightweight enough to not consume more coordination bandwidth than the alignment is worth.
The Verdict on Measure What Matters
A useful book that could have been a great one. The core framework — OKRs as a focus and alignment tool — is genuinely valuable and Doerr explains it clearly. The Intel and Google case studies are excellent. The CFR companion system is underemphasized but important.
The weakness is the book’s treatment of its own context. The OKR cases are almost entirely from Silicon Valley tech companies with exceptional talent, strong intrinsic motivation, and clear metrics for success. The adaptation required for other contexts is largely left to the reader. The failure modes section is too brief and too charitable — most OKR implementations fail, and understanding why is as important as understanding the system itself.
Read it alongside Grove’s High Output Management — the source material for OKRs — to get the underlying philosophy rather than just the implementation guide. For the personal performance dimension of goal-setting and accountability, our piece on grit and resilience explores why sustained commitment to specific goals matters more than the goal-setting system itself. The psychological foundation of the accountability practices OKRs require is addressed in our mental toughness training guide. The combination is significantly more powerful than either book alone. Grove explains why clear goals and measurement matter at a level of rigor Doerr doesn’t match. Doerr provides the implementation detail and modern case studies that Grove’s book, written in 1983, obviously lacks.
Reader Questions About Measure Matters Summary
What does OKR stand for and how does it work?
OKR stands for Objectives and Key Results. An objective is what you want to achieve — a significant, inspiring goal. Key results are how you measure whether you’re achieving it — specific, time-bound, measurable outcomes. Together they create a feedback loop: the objective gives direction, the key results tell you if you’re moving in that direction.
Where did OKRs come from?
Andy Grove developed the system at Intel in the 1970s as a refinement of Peter Drucker’s Management by Objectives framework. John Doerr learned it from Grove in 1975, brought it to Kleiner Perkins, and introduced it to Google in 1999. Google has used OKRs continuously since then.
Why doesn’t Google score 100% on its OKRs?
By design. Google targets approximately 70% of its key results as success. The reasoning: if you consistently hit 100%, your objectives aren’t ambitious enough. Stretch goals that are somewhat beyond reach push teams to perform at higher levels than comfortable goals would. A 70% hit rate on ambitious goals produces better outcomes than a 100% hit rate on safe ones.
Should OKR scores be tied to compensation?
No, and Doerr is explicit about this. When OKR scores determine bonuses, people stop setting stretch goals and start sandbagging — setting targets they’re certain they’ll hit. The stretch goal value disappears. OKRs should inform performance conversations but not mechanically determine compensation decisions.
How many OKRs should a team have per quarter?
Three to five objectives, with two to five key results per objective. The constraint is intentional — the focus function only works if the list is short enough to be meaningful. More than five objectives per quarter is usually a sign that the prioritization work hasn’t been done.
What is the most common reason OKR implementations fail?
Absence of the check-in cadence. OKRs are set, then forgotten until the end-of-quarter review. The feedback loop that makes the system valuable — regular monitoring of progress, course correction when you’re off track, learning from the data — requires frequent check-ins. Without them, OKRs are just an annual goal-setting exercise with fancier language.
What are CFRs and why do they matter?
Conversations, Feedback, Recognition — the human management complement to OKRs. OKRs measure organizational progress. CFRs develop the people responsible for that progress. Without CFRs, OKRs produce measurement without development. The two systems work together: OKRs tell you what’s happening, CFRs build the capacity for it to go better.
Can OKRs work for small businesses or is it only for large companies?
The framework is scale-independent. A five-person business can benefit from the focus and alignment functions of OKRs. The implementation details need significant adaptation from the Google/Intel context — the formal quarterly review process, the company-wide transparency system, the stretch goal culture all require adjustment for smaller organizations.
What’s the difference between OKRs and KPIs?
KPIs (Key Performance Indicators) measure ongoing operational performance — metrics you track continuously to know if the business is healthy. OKRs measure progress toward specific objectives over a defined period. KPIs tell you if your existing operations are working. OKRs tell you if you’re achieving new goals. Both are useful; they serve different functions.
How does Measure What Matters compare to Andy Grove’s High Output Management?
Doerr’s book is more accessible and provides more modern case studies. Grove’s book is more philosophically rigorous and explains the underlying management theory at a deeper level. Grove explains why measurable goals and output management matter. Doerr explains how to implement one specific system built on those principles. Reading both together is significantly more valuable than reading either alone.
Related: Attached Summary
Related: Give and Take Summary
Related: Unlimited Power Summary
Related: 12 Rules for Life Summary
Related: The Black Swan Summary
Related: Bigger Leaner Stronger Summary
Related: Brave New World Summary
References
Editorial StandardsCorrectionsMedical DisclaimerAbout Our ContentAffiliate DisclosureSite Map