A Goal You Can't Grade Is Just a Mood

In November 1979 a sales manager at Intel sent head office a message nobody wanted to read. Motorola's new chip was beating theirs, and customers knew it. Intel answered with a campaign it called Operation Crush. One clear objective, a few hard numbers underneath it, and a target of 2,000 design wins in a year. They finished well past it.

The system behind that campaign belonged to Andy Grove, who ran Intel. Twenty years later a man who had learned it working at Intel in the 1970s, John Doerr, stood next to a ping-pong table at a startup of thirty or so people and taught it to the founders. The startup was Google. The system was OKRs. Objectives and key results.

It is easy to hear that and switch off. Intel. Google. Thousands of staff and layers of managers. What does any of it have to do with one venue, a team of twelve, and an owner doing the roster on a Sunday night?

More than you would think. Strip away the software, the cascading and the offsites, and OKRs answer the one question most owners dodge. At the end of the next three months, how will you know if it worked?

Most of us can't answer it. We have a goal. Have a big Christmas. Grow the catering. Get more regulars in. Then the quarter ends and the goal gets graded by feel. It was a good Christmas. Sort of. Busy, anyway.

A goal you can't grade is just a mood.

What OKRs Actually Are

Two parts. That is the whole framework.

The objective is what you want. It is written in words, not numbers, and it should make you want to get out of bed. "Become the coffee the business park runs on." "Own Christmas catering in our suburb." Short, specific and a bit ambitious.

The key results are how you will know you got there. Numbers, with a date on them. Usually three. Doerr boils the whole thing down to one sentence you can fill in: we will do this objective, as measured by these key results.

Here is the test that makes it work. At the end of the quarter, each key result was either hit or it wasn't. No "kind of". If you and your manager could have a genuine argument about whether it was hit, it isn't a key result yet.

And here is the rule nearly everyone gets wrong the first time. A key result is an outcome, not a task. "Post on Instagram three times a week" is a task. You could do it perfectly for twelve weeks and nothing would change. "Twelve catering enquiries from Instagram by December" is a result. Tasks are how you get there. They live on your to-do list, not on the OKR.

We have written before about Ray Dalio's point that a goal belongs in numbers, not feelings, in The System That Runs Our Entire Business. OKRs split that goal in two, the why and the proof, and then add two rules that stop the numbers lying to you. We will get to those.

OKRs Cut Down for One Venue

The big company version is built for layers. Company OKRs at the top, department OKRs underneath, team OKRs under those, all rolling up into each other and tracked in software. Several objectives per team, each with its own handful of key results.

You don't have layers. You have you, maybe a manager or a head chef, and a team on the floor. So cut it right back.

One objective a quarter. You have one set of eyes and one set of hours. Two objectives in a single venue usually means neither gets done properly.

Three key results. Enough to define done from a couple of angles. Few enough that you can say them without looking.

One sheet of paper. On the wall out the back, where the team walks past it every shift. Not in a spreadsheet on your laptop. The whole point is that everyone knows what winning looks like this quarter.

Let the team write one. Doerr pushes hard on this. Goals handed down get compliance. Goals people helped write get effort. Tell your floor manager the objective and ask what number would prove you got there. Their answer is often sharper than yours, because they are the ones standing at the pass all day.

If you have already worked out the one ratio that drags your whole business up, the one we wrote about in Good Is the Enemy of Great, that is where your key results should come from. The ratio is the direction. The OKR is this quarter's push on it.

And a quarter is the right length. Long enough to move something real. Short enough that you can't pretend the deadline is ages away. The quarter that runs into Christmas makes a good first one, because catering and end of year functions give you something real to aim at.

Pair Every Key Result, and Aim to Miss

Two rules separate an OKR from a target scribbled on a whiteboard.

Pair every number. Grove wrote about this years before Doerr did. Push hard on one number and something else quietly breaks. So every key result that pushes gets a partner that watches for the damage. He called it measuring the effect and the counter-effect.

In a cafe it looks like this. These are examples, so swap in your own numbers.

• Catering orders up to $3,000 a month, paired with Saturday ticket times staying under eight minutes. The same kitchen preps both, and your regulars will notice before you do.
• Average spend up a dollar, paired with regulars' weekly visits holding steady. Upsell too hard and the people who pay your rent start coming in less.
• Food waste down by half, paired with nothing in the cabinet selling out before 1pm. The fastest way to cut waste is to run out of food, and that is not the win you were after.
• Four new office accounts ordering every week, paired with all four still ordering in week twelve. Signing them is easy. Keeping them is the business.

Without the pair, the number wins and the business loses. With it, nobody on the team can hit the target by quietly breaking something else.

Aim to miss. Doerr splits OKRs into two kinds. Committed ones you have to hit, like getting the new oven in and working before the Christmas rush. Aspirational ones are meant to stretch you. On those, landing around 0.7 out of 1 counts as a good result. If you hit every stretch goal in full, you set them too low.

That is a big shift for owners who read a missed target as failure. Say the target was six new office accounts and you landed four. You didn't fail. You landed four accounts you would never have chased if the target had been two.

James Clear says you fall to the level of your systems, and we agreed with him in You Don't Rise to Your Goals. OKRs don't argue with that. They just tell you which system is worth building this quarter.

The Exercise: One Sheet, Ten Minutes a Week

This week, take one A4 sheet and write it out.

The objective. One sentence, in words, that you would be proud to have done by the end of the quarter.

Three key results. Numbers with a date. Then check each one against three questions. Is it an outcome and not a task? Could anyone argue about whether it got hit? Does it have a pair watching for damage?

Now sit down with whoever runs your floor and ask them to rewrite one of the three. Use theirs.

Pin the sheet where the team sees it every shift. Every Monday, spend ten minutes on it. Give each key result a score out of ten for how likely you are to hit it, as of today. When a score drops, that is your cue to talk about it. Week four, while there is still time to fix it. Not week thirteen.

At the end of the quarter, grade each key result from 0 to 1. Write one line on what you learned. Then write the next sheet.

By the second quarter you will notice the conversations out the back change. Less "are we busy?" and more "are we on track?" That is what a goal you can grade does. It turns a mood into a number the whole team can read.

---

Further reading: Measure What Matters by John Doerr

---

Inspired by Measure What Matters by John Doerr. If this hit home, the book goes deeper.

Keep reading: Good Is the Enemy of Great. Your Cafe Is Proof. · The System That Runs Our Entire Business

If you want a Perth coffee roaster who thinks about your business the way we write about it, let's talk.