Busy isn’t an outcome. Neither is shipping.
Introducing OKRs inside Canada’s public broadcaster meant putting one uncomfortable question to every goal. What would have to measurably change for us to say this worked? Most goals couldn’t answer it.
8
Rewrites of one quarter’s goals in two weeks
2×
Where the final targets landed, vs. the comfortable ones
0
Goals allowed without a baseline and a number
CBC Digital ran the digital side of Canada’s national public broadcaster. Product, engineering, design and content, at massive national scale, with more initiatives underway than anyone could hold in their head at once. In 2017 I was in a technology leadership role there, and I helped introduce Objectives and Key Results, OKRs, to change how goals, priorities and measurement worked.
Big organizations never run short of activity. The hard part is knowing whether any of it’s moving an outcome that matters, and at that size it’s wild how long you can go without finding out.
So the real change was one question every goal had to answer: what would have to measurably change for us to say this work actually succeeded?
- Organization
- CBC Digital, Canada’s public broadcaster
- Year
- 2017
- Role
- Technology leadership
- The work
- Introducing Objectives & Key Results · outcomes, not activity
“Busy teams plus a delivered plan equals progress. Measure the work, what shipped, launched and completed, and the outcomes will follow.”
“Nothing counts until you can say what would have to measurably change. Written honestly, most goals couldn’t say.”
The first thing I found was how easy it is to mistake activity for results. The goals on the wall all sounded reasonable. None of them could tell you whether anything had been accomplished:
- “Improve morale.” Improve it from what, to what, by when?
- “Encourage more training.” Measured how, and against what participation today?
- “Grow the product.” How many additional users, over what period?
- “Make the experience better.” Better than what baseline, by how much?
Every one of those lets a team work hard for a quarter and declare success without anything measurably changing. That’s the default setting of planning, and a goal with no baseline’s dead on arrival, because there’s nothing it can be measured against and so nothing it can miss.
That sounds obvious once it’s written down. Fixing it changes a lot, and the first thing you notice is how uncomfortable the honest version of a goal feels.
- 01
Make every goal falsifiable
Each objective had to name its baseline, its target and its date. A goal that can’t fail is a slogan.
- 02
Push targets past the comfortable
People naturally commit to what they know they can deliver. That’s sensible from where they sit, and it makes project plans comfortable and growth targets meaningless. In one planning cycle I rewrote my own OKRs roughly eight times in two weeks, doubling some targets in the final iterations. I was hunting for the line between predictable execution and genuine ambition. If you already know exactly how you’ll hit a target, there’s no magic in hitting it, and it isn’t forcing anyone to think differently.
- 03
Define the outcome first, then chase the dependencies
At CBC’s scale, meaningful outcomes rarely belonged to one team. A goal might need product, engineering, design and content all to change something. Find that out at delivery time and you’re screwed, because the other teams already planned their quarter around something else. So the discipline was to hold the ambition and not negotiate it away early. Define the outcome that matters, work out what it depends on, then socialize it with the teams involved and find where their objectives intersect. That made OKRs more useful than a scorecard. They exposed the places where organizational dependencies would quietly stop an important outcome from ever happening.
| Before | After | |
|---|---|---|
| A goal looked like | “Make the experience better” | A baseline, a target, and a date |
| Targets were set by | What teams knew they could deliver | Rewritten until they demanded new thinking |
| Cross-team dependencies | Discovered at delivery time | Exposed at planning time |
What measurably changed at CBC was the goals themselves: what a goal was allowed to look like, how hard it got pushed, and when its dependencies surfaced. I’ll be candid about the rest. The downstream organizational outcomes from that shift weren’t tracked in a way this page could honestly cite, so it doesn’t cite them.
What the work proved was the discipline, and I’ve carried it into every engagement since. The question I care about is whether something important changed.
CBC Digital and the role are Matt’s own history. The numbers on this page are about how goals got set. The downstream organizational outcomes weren’t tracked to this page’s standard, so this case study doesn’t invent them.
When I look at a business today, the number of initiatives underway interests me far less than four simpler questions. What are you trying to change? How will you know if it changed? What’s stopping it from changing? And is the work eating your team’s time actually connected to that result? Often the fastest route to growth is naming the few outcomes that matter and stopping the work that doesn’t move them. That second part is the one nobody wants to do.
What’s all the work eating your team’s time actually changing?
The Next Version finds where the money is actually stuck (starting with whether your numbers are even real) and names each move, what it’s worth, and who runs it. You keep the roadmap either way.