The Reading Instrument
A structured way to read your organisation's OKR practice as evidence of what your management system actually is.
This is an instrument, not a survey. It produces conversation, not a score. Two people using it on the same organisation may reach different readings, and the disagreement is part of the value - the places where readings diverge are usually the places where the system is least clear about itself.
Use this in a workshop, a leadership conversation, or a piece of discovery work. The instrument is not designed for individual completion; the value is in the discussion the questions provoke.
How to use it
For each section, the question is not what is the right answer. The question is what does your honest answer tell you about the system. Where the honest answer makes you uncomfortable, that is where the diagnostic information is.
Work through the four sections in order. The signals build on each other. By the end, the pattern of answers should give you a reading of which of the four signals your system is most clearly producing, and which of the three constraints is most clearly missing.
Take notes. The reading is more useful with a record than without.
Section one - what the artefact carries
Pick a current OKR in your organisation. Not the best one or the worst one. A typical one.
- Who wrote this Objective? Was it written by the team that has to deliver it, or written for them?
- If someone outside the team reads this Objective without context, does it make the team's intent legible to them? Or do they need an explanation?
- What does the Objective imply about what the team will and will not do? Has any work been refused on the basis of this Objective in the last quarter?
- If the team hit all the Key Results, would anything important have changed in the world for the users of this team's work? Or would the team simply have done some things?
The pattern in your answers is the first signal of what your system is for.
Section two - what happens when a Key Result is missed
Think about the last time a KR in your organisation missed its target by a meaningful margin.
- Was the miss treated as a problem to explain, or as information to understand?
- Whose name was on the miss? Did the miss appear in any individual's performance documentation?
- Was the miss compared to a target, or to a range? Did anyone ask whether the variation was special-cause or common-cause?
- What changed as a result of the miss? A target? A process? A person's prospects? Nothing?
The answers describe what your system believes about variation, accountability, and human dignity. The vocabulary may not be in the answers, but the philosophy is.
Section three - how OKRs live or die between quarters
Think about a typical week in your organisation, six weeks into a quarter.
- In how many recurring meetings - standup, sprint review, planning, one-to-ones, leadership reviews - does an OKR get mentioned by name?
- When backlog work is added or removed, is the reasoning referenced to a Key Result?
- Have any of the OKRs been revised since they were written? Have any been killed? On what grounds?
- If the OKRs vanished tomorrow, would anyone notice? Within what timeframe?
The answers describe the relationship between the OKR artefact and the operating rhythm of the work. The smaller the relationship, the more the artefact is doing none of its intended work.
Section four - what is sayable and what is not
This is the hardest section. Take it slowly.
- Can a team in your organisation say "this corporate Objective does not apply to us" without political cost? What happens when they try?
- Can a team commit to a stretch Key Result, miss it, and have the conversation be about what was learned rather than about who is to blame?
- Can a team change a Key Result mid-quarter because they have learned something, without it being read as failure of planning?
- When the senior team in your organisation reports its own OKR scores, how honest are those scores? When did a senior leader last publicly report a 60% on a stretch Objective?
The answers describe the safety constraint. Where the answers are no, that is not possible here, the system is operating under absent safety - and most of the other constraints will be partially defeated regardless of what else you do.
Reading the pattern
Look across the four sections. Three patterns to look for.
Which signal is most clearly present. Where your answers cluster around the artefact carrying someone else's intent and missing KRs being personally consequential, the signal is used-as-a-stick and used-for-control. Where your answers cluster around OKRs being absent from the operating rhythm and nothing changing when they are missed, the signal is set-and-forgotten. Where your answers cluster around OKRs being cited but disconnected from decision-making, the signal is used-and-ignored.
Multiple signals can be present at once. They often are.
Which constraint is most clearly missing. If section four produces the most uncomfortable answers, safety is the missing constraint. If section three is the worst, rhythm. If section one shows that intent is not radiating and section two shows that the system treats every miss as a story without statistical literacy, visibility is missing.
The missing constraint is the place to start. Adding constraints in the wrong order does not work - building rhythm without safety produces forced cheerful theatre, building visibility without safety produces surveillance.
What philosophy is being expressed. This is the question that the conversation should keep returning to. The pattern of answers does not just diagnose; it reveals the worldview that the OKR system is operating under. The worldview is usually not the one anyone would defend out loud.
The reading is most useful when it makes the worldview visible. That visibility is what allows the question - is this the philosophy we would choose if we were choosing deliberately? - to be asked at all.
What this instrument does not do
Three things to be clear about.
This instrument does not give you a score. The pattern of answers is qualitative. Two organisations might produce the same set of answers and need different interventions, because the underlying systems are different in ways the instrument cannot see.
This instrument does not tell you what to do. Diagnosis precedes intervention. Once the reading is complete, the constraints section of the playbook describes what kinds of intervention are likely to help. The instrument tells you where to start.
This instrument does not work in isolation. It works in conversation. A leadership team using it together, working through the questions and disagreeing about the answers, will produce a more useful reading than any individual completing it alone. The instrument is designed for disagreement. The places where readings differ are the places where the system is least settled.
An artefact of the OKRs playbook. Profound - a collection of playbooks and instruments for understanding a system of work.
Licensed under CC BY-SA 4.0.
Last updated 15 May 2026
Licensed under CC BY-SA 4.0.