IX Coach

OKR Coaching: What an OKR Coach Does, Where OKRs Go Wrong, and When You Need One

What is OKR coaching?

OKR coaching is help writing, running, and reviewing Objectives and Key Results, usually for a team and sometimes for one person. This guide covers where OKRs came from, what an OKR coach actually does, the failure modes practitioners describe most often, how team and individual coaching differ, what the research does and does not show, and how to tell whether you need a coach at all.

OKR coaching is help with Objectives and Key Results, the goal-setting method in which a short, ambitious objective is paired with a few measurable key results that show whether you got there. A coach in this area helps a team or a single person write good OKRs, line them up with what the organization is trying to do, run the regular check-ins, and hold an honest review at the end of the cycle. The coaching is mostly about the conversations around the framework rather than the framework itself, since most people can learn the format in an afternoon and still struggle to use it.

This page covers where OKRs came from, what an OKR coach does in practice, the failure modes that come up most often (each one attributed to where it comes from, or marked as common practitioner guidance), how coaching a team differs from coaching one person through personal OKRs, what the research says and where it runs thin, and a short way to decide whether you need a coach at all.

What OKRs are, and where they came from

An OKR has two parts. The Objective says in plain words where you want to go. The Key Results are the measurable evidence that you are getting there. John Doerr's OKR site describes the usual shape as one objective with roughly three to five key results beneath it, and Google's published guide recommends picking three to five objectives with around three key results each.

The method is credited to Andy Grove, who ran Intel and described the approach in his 1983 book High Output Management, framing it as two questions: where do I want to go, and how will I pace myself to see whether I am getting there. Grove's own name for it at Intel was iMBOs, short for Intel Management by Objectives, building on Peter Drucker's earlier Management by Objectives. John Doerr, who worked at Intel in the 1970s, learned it from Grove, and in 1999 introduced it to Google as a venture investor at Kleiner Perkins. His 2018 book Measure What Matters, published by Portfolio/Penguin, is what carried the method outside the technology industry.

Two things in that history are worth holding onto. The first is that OKRs are a management practice that grew up inside companies, not a finding from a laboratory. The second is that Doerr's site describes OKRs as divorced from compensation, which it presents as a difference from the older Management by Objectives approach, and Google's guide likewise says OKRs are not a way to evaluate an individual. That point comes back in the failure modes below.

Explore: okrs

What an OKR coach actually does

We are not aware of a licensing body or single agreed standard for the job, so the following is a description of what practitioners commonly do rather than a certified definition, and the mix differs by coach. Four activities come up most often.

Drafting. Much of the early work is editing: turning a list of projects into an objective someone could picture achieving, and turning vague aims like "improve onboarding" into key results a team could actually score. A coach tends to ask the question that exposes the gap, such as what would be different in the world if this were done, and how anyone would know.

Alignment. In a team or company, objectives at different levels are meant to connect. A coach helps people see where their objective supports someone else's, where two teams are quietly competing for the same outcome, and where nobody owns something important. This is a conversation across people more than a document edit.

Check-in cadence. Google's guide describes setting annual and quarterly OKRs and grading them at the end of the quarter, with teams revisiting them several times during it. Many practitioners add a short, regular check-in, often weekly, to ask what has moved, what is blocked, and whether the objective still makes sense; the weekly rhythm is common practitioner guidance rather than something we have a single source for. A coach often helps establish the rhythm and keeps the meeting about decisions rather than status recitation.

Retrospectives. At the end of a cycle, a coach helps the group score honestly and ask what the results say about the objectives, as opposed to who failed. Google's guide grades key results on a scale from 0.0 to 1.0, so the review is partly a conversation about why a score landed where it did.

The failure modes practitioners describe most

Too many objectives. The point of the format is focus, and a long list defeats it. Google's guide recommends three to five objectives. A coach's contribution here is usually helping people choose, which is harder than adding, since every item on the list has someone who cares about it.

Tasks disguised as key results. Christina Wodtke, author of Radical Focus (2016), argues that key results should describe outcomes rather than activities. In an interview on the Weekdone blog she gives "launch a new CRM tool" as a task and asks "so what?" of it, pointing instead at the result the launch is supposed to produce, such as more sales closed. The reasoning, as we would put it, is that a key result written as a launch can be completed on schedule while the number it was meant to move stays where it was.

Sandbagging versus stretch. Google's guide says OKRs are meant to be ambitious, treats scores around 0.6 to 0.7 as the expected sweet spot, and says that someone who consistently scores 1.0 probably is not setting ambitious enough goals. Doerr's site separates aspirational OKRs (stretch goals or moonshots) from committed ones that are expected to land. The 0.6 to 0.7 figure is a Google convention, not a research result, and other organizations calibrate differently. Sandbagging, setting a target you are sure to hit, is the failure this convention is designed to counter.

Tying OKRs to compensation or performance reviews. Google's guide states that OKRs are not a comprehensive means to evaluate an individual, and Doerr's site describes OKRs as divorced from compensation. Wodtke's stated concern, in the same interview, is that linking pay to completion pushes people toward goals they can be sure of finishing. A related academic argument comes from Ordóñez, Schweitzer, Galinsky and Bazerman (2009), who reviewed side effects of over-prescribed goal setting, including narrow focus on the measured target, more risk-taking and unethical behavior, and reduced intrinsic motivation. That paper is about goal setting generally, not OKRs, and it is a position and review paper rather than a trial of any method.

Setting and forgetting. A common pattern is writing OKRs with energy in week one and reading them again at the end of the quarter. Researchers who interviewed engineers at one large software company, in a 2023 preprint by Butler, Zimmermann and Bird, reported that tracking, measuring and setting goals is hard work whatever tools are used, and that translating executive-level direction into actionable work is a recurring difficulty. That is one company, so treat it as an illustration rather than a rate.

Individual OKRs for everyone. According to Wikipedia's OKR entry, OKRs set at the level of individual contributors tend to read like task lists and get conflated with performance reviews, and Rick Klau of Google Ventures advised skipping individual OKRs, especially at younger and smaller companies, in favor of company and team levels. That is one practitioner's position, reported secondhand, but it is a useful prompt to ask who an individual objective is actually for.

Explore: stretch goals · process vs outcome goals

Coaching a team versus coaching one person

The two are close to different jobs. What follows is reasoning from how the method is built, not a research finding. With a team, the hard part is usually between people: agreeing which objective matters most when two reasonable leaders disagree, naming a key result that more than one person controls, and making it safe to report a low score. Scoring honestly depends on people being willing to say that something did not work, which is part of why teams working on psychological safety tend to find it relevant here. The Butler preprint above also highlights middle managers as the link between vision and day-to-day work, so a team coach often spends real time with them.

With one person doing personal OKRs, there is no alignment to broker, and the conversation moves inward. The questions become what you actually want from this quarter, whether the objective is yours or one you inherited from someone's expectations, and what gets in the way of looking at the numbers. Goals a person has chosen for their own reasons are the subject of a separate line of research on self-concordant goals, and personal OKRs tend to work better when the objective is truly the person's own. The main risk is a mirror image of the team case: with no one else to answer to, a quarter can pass with no check-in at all, which is where an accountability partner, a coach, or a regular reflective conversation does the work.

Explore: accountability partners · psychological safety teams · self concordant goals

What the research does and does not say

The evidence behind OKRs splits into two layers. The first is goal-setting research. Edwin Locke and Gary Latham summarized decades of studies in a 2002 article in American Psychologist, and the central finding they describe is that specific, difficult goals tend to lead to higher performance than easy goals or an instruction to do your best, provided people get feedback on progress, are committed to the goal, and have the ability and knowledge to do the task. The ambitious-but-measurable shape of an OKR is consistent with that.

The second layer is OKR-specific outcome research, and it is thin. A 2024 systematic mapping study by Silva and Santos found 47 academic sources on OKR use and concluded that OKR use is under-documented from a theoretical point of view, with few studies addressing the topic in depth. We did not find a controlled study showing that OKRs, or coaching on OKRs, improve organizational results compared with other goal-setting approaches. Broader evidence exists that workplace coaching is associated with positive individual outcomes, for example a 2014 meta-analysis by Theeboom, Beersma and van Vianen, but that work is about coaching in general and does not test OKR coaching against anything.

So the honest reading is that the underlying principles have support, the packaging has mostly practitioner testimony behind it, and the combination of an OKR coach has not been tested head to head. That is a reason to treat any confident claim about OKR coaching results, including from coaches, with some care.

Explore: goal setting theory

How to tell whether you need a coach at all

Many teams do fine without one. A few questions can help you locate where yours stands, and none of them is a diagnosis.

Could each person on the team state the current objective without looking it up? If not, the issue is probably focus or visibility, and cutting the list may help more than adding a coach. Are your key results things you can measure by looking at a number, or are they things you can complete? If they are mostly completions, a rewrite pass, which a colleague who has not seen the draft can often do, may be enough. Do scores come in at 1.0 every quarter, or at 0.3 every quarter? Either pattern suggests the targets are calibrated to something other than the work. Is anyone nervous about reporting a low score? If so, the problem may be how the scores are used, and it is worth checking whether OKRs have drifted into performance reviews.

A coach tends to be most useful when the difficulty is between people, when two groups will not agree on priorities, or when a leader wants an outside voice in the room. It tends to be less necessary when the difficulty is that nobody has read the OKR document since week two, which a calendar invite and a ten-minute agenda can often address. Doing one cycle on your own, scoring it honestly, and noting where you got stuck also gives a coach, if you hire one later, something concrete to work from.

Explore: smart goals

Where IX Coach fits

IX Coach is an AI coaching system you talk to directly, designed around reflection, goal clarity, and the move from insight to action, and it tells you it is an AI. It is not an OKR platform or a team rollout program, so it does not run a company's OKR cycle. Where it may be relevant is the one-person end of this page: thinking through what a personal objective is really for, noticing when a key result is a task in disguise, or having a conversation each week about what moved. You can try it for seven days at no cost.

Research

Practice this with IX Coach

Try this practice

Keep reading