The OKRs are rarely the problem.
Your goals went quiet in March.
The infrastructure around them is — and that's what we install.
You wrote the objectives. You assigned the owners. Somewhere around the second month the updates stopped, and by the end of the quarter nobody could tell you where anything stood. That is the ordinary outcome, and it is almost never caused by the goals being badly written. It's caused by there being no scorecard, no review rhythm, and nobody whose job it is to run the program.
The problem
Almost every company we talk to has done this once already. The goals went up on a slide in January, everyone nodded, and the whole thing was quietly dead by spring. What usually gets blamed is the framework, or the team's discipline, or the fact that the quarter got busy.
It's rarely any of those. Goals are not self-executing. They need someone who owns the program, a place where progress is visible without anyone assembling it by hand, a rhythm that pulls attention back to them, and numbers that show up on their own. Most companies write the goals and skip all four, then conclude that goal-setting doesn't work here.
There's a second version of this problem, and it needs the opposite fix. Some companies have the rhythm and the reporting, and the goals themselves are the weak part — too many of them, no baselines underneath the targets, key results that don't add up to the objective they sit under. Fixing infrastructure for that company is wasted money, and rewriting goals for the first one is wasted money. Telling the two apart is where this starts.
The diagnosis
Which one do you have?
Objectives break in two distinct ways, and the intervention is completely different depending on which one you're looking at. Read both lists. Most leadership teams recognize themselves in one of them within about four bullets.
A quality problem
The objectives themselves are weak.
- Too many objectives, which creates the appearance of strategy without any actual prioritization.
- Objectives that describe business-as-usual rather than meaningful change.
- Key results that don't add up to the objective — loosely related activity standing in for a coherent set of measures.
- Targets set without baselines, so the company is measuring something it has never measured and the target is a guess.
- Objectives change frequently without anyone acknowledging it, so the organization chases new things before finishing old ones.
- People describe the company's goals differently depending on who you ask.
An infrastructure problem
The objectives are fine. Nothing makes them run.
- Objectives live on a slide or in a spreadsheet and don't influence a single weekly decision.
- Objectives were assigned to owners, but nobody defined what owning one actually means.
- There is no cadence for reviewing progress — goal-setting was a kickoff exercise that faded.
- There is no artifact for tracking. No scorecard, no status, no place where progress lives.
- Nobody owns the program itself — no one running the calendar, no one reinforcing the message.
- The numbers depend on data sources with lag times, and the review rhythm was never designed around those dependencies.
- Teams set their own objectives with no connection to company-level goals.
If you can't tell which list is yours — or you see yourself in both — that is the ordinary case, and it is exactly what the diagnostic is for. Nobody should be quoting you an install before that question is answered.
What we install
Most OKR consultants leave you a document. We leave a system that produces the number.
That means the roles first, briefed and accepted out loud rather than assigned on a slide: someone at or near the top who writes their own objectives and grades them honestly, someone running the calendar, an owner for every key result, and a named owner for every number — including the ones that come from your accountant or a third-party system. That last one is the dependency nobody designs for, and it is the most common reason a review meeting turns into a status-chasing meeting.
Then the scorecard, built as a real instrument rather than a spreadsheet somebody maintains on Sunday nights. Wired into the tools your work already lives in, with the data pull automated wherever the data owner can be a system instead of a person. This is the part that separates us from the rest of the category, and it's the rest of our practice pointed at this one problem.
Then the rhythm, designed into meetings that already exist. We don't add a standing meeting to your calendar. A goal system that requires one has already lost, because the new meeting is the first thing cut when the quarter gets busy.
If you set goals this year and can't say where they stand right now, that's a 30-minute conversation.
How it works
A diagnostic first, and it's paid. Two weeks. We run the readiness conversation with you and two or three of your leaders, then audit every goal you already have — metric definition, data source, current baseline, reporting lag, for each one. Anything that can't fill in all four columns gets flagged. You get a named data owner for every number, decision rights written down for when the room can't agree, and a written recommendation: ready, not ready, or fix this first.
Stop there and you own the map. That's the honest version of what the fee buys, and it's the courting period for both of us.
Start here
Goal diagnostic
$10K Two weeks
Stop here and you keep everything — the readiness findings, the baseline audit, the named data owners. Keep going within 30 days and the $10K comes off the install. The diagnostic is also what tells us which depth below actually fits.
Then you pick the depth
Activation
$30K
One quarter
For goals that exist and never ran. Roles accepted, baseline gaps fixed, the scorecard built, the rhythm designed into your calendar, and us in the room for the first 90 days of tracking.
First cycle install
$60K
One quarter
Everything in Activation, plus the objectives crafted with your leadership, cascaded across teams, launched, and coached through a full cycle to grading. Ends with your team writing cycle two while we're still in the room.
Full cadence install
$150K
Two cycles, six months
Everything above, across company and team layers, with owners trained in each team and two cycles graded. The scorecard wired into your systems so the numbers arrive without anyone typing them in.
Fixed fees, agreed before we start. The diagnostic credits against any install begun within 30 days.
What we need from you: a champion at or near the top who writes their own objectives and grades them honestly, a program owner with an hour a month, decision rights settled before we start, goals decoupled from compensation for at least the first two cycles, and attendance at the working sessions.
Each of those is a real filter rather than a formality. A company that won't agree to them is a company whose program was going to fail, and we'd rather find that out now than in month four.
What finished means
We don't consider an OKR program installed until your team runs a cycle without us in the room.
Not a set of goals in a document, and not one quarter that we facilitated. The test is whether it still happens when we're gone.
Full terms in the agreement.
When we'll tell you not to
Three things make this fail no matter how good the install is. If any of them is true, we'll say so on the first call rather than after you've paid us.
- No executive sponsor. If nobody at the top is genuinely behind it, don't start. We won't take the work either.
- No strategy. Objectives translate a direction into action. If the direction isn't settled, that's a different engagement and it has to come first.
- Goals tied to compensation. We'll ask you to decouple them for at least the first two cycles. If that isn't on the table, the tracking data will be dishonest and the program becomes theater.
Common questions
We tried this before and it died. Why would this be different?
Because that's the normal outcome, and it usually isn't the goals that failed. Most programs die from missing infrastructure — no scorecard, no cadence, no one who owns the program — and rewriting the objectives does nothing about any of that. The diagnostic tells you which of the two problems you actually have before anyone proposes a fix.
How long before something is actually running?
Locked objectives and a live scorecard inside three weeks. That is the standard we hold ourselves to, and it is deliberately not months of crafting sessions with nothing committed at the end.
This is one more thing my team doesn't have time for.
The person running the program spends about an hour a month, and the tracking takes fifteen minutes inside a meeting that already exists. If a goal system needs a new standing meeting to survive, it was designed wrong.
What happens to the goals we already wrote?
We audit them rather than replace them. Every existing key result gets four columns filled in — metric definition, data source, current baseline, reporting lag — and the ones that can't fill in all four get flagged and triaged. Often the goals are fine and the measurement underneath them was never built.
Should we tie OKRs to compensation?
Not for the first two cycles, and we'll ask you to commit to that before we start. When the number decides someone's bonus, people set targets they know they can hit and grade themselves generously. The tracking data stops being honest, and honest data is the entire point.
Do we have to use OKRs specifically?
No. Some companies need a simpler prioritization framework, and we'll say so if that's what we find. The operating system underneath — owners, a scorecard, a review rhythm, someone running it — is what makes goals work, and that part doesn't change with the framework you put on top.
What if the diagnostic says we should not do this?
Then we'll tell you, and you keep everything we produced — the readiness findings, the baseline audit, the named data owners. You spent $10K finding out what was worth doing. We'd rather lose the install than sell you one that was going to fail.
More of what we do
Custom software
You’ve outgrown spreadsheets and off-the-shelf tools. We build software your team will actually use, and you own all of it.
Read more →Due diligence
A buyer’s lens on your business before a buyer takes one — findings ranked by valuation risk.
Read more →AI solutions
Seats aren’t adoption. A paid workflow audit, then the expensive work rebuilt with AI inside it.
Read more →Financial models
Stop deciding on last month’s P&L. Forward-looking models with scenarios you can test before you commit.
Read more →Ob — Objectives
Writing on Objectives
Objectives is one of the twelve elements of Harmonic Architecture. These are the issues the rest of the archive keeps pointing back to.
- Speed without direction gets you nowhereSpeed without direction is just motion. Velocity is the number that actually matters.
- What needs to be true?The question that sidesteps "that's impossible" and gets a team building momentum.
- Design a Future That Gives You GoosebumpsStrategic planning applied to a team of one. Part 1: turning a vision into goals.
Find out which problem you have.
A 30-minute conversation about the goals you set, what happened to them, and whether the fix is the goals or the system around them. If we're not the right help, we'll tell you.
Not ready to talk?
Get the Catalyst — Kevin's weekly newsletter on designing businesses that scale without depending on you.