Meet Dave - a response to the stick

A measure is a mirror – until it becomes a lever

She ended the call with a sentence that was both unsettling and deeply mischievous: "I want a stick to hit them with."

Not surprising, exactly – I'd seen it coming – but it landed uncomfortably all the same, right there at the end of one of the more trying meetings I've sat through. The subject was one close to my heart: measuring teams, ranking them, and finding a single number to hold them all to account with. I say close to my heart because it's actually the entire opposite of my stance.

Her plan was carefully made and, I've no doubt, well meant. She wanted to take the metrics that teams already produce in abundance, build a scoreboard across all hundred-plus of them, and rank them top to bottom. The teams at the bottom would then, in her words, be hit with the stick – a rather more aggressive version of "having a word." I mean, come on, the real question for teams who may be struggling is "how can I help?", but let's not get into that.

I cautioned against it, at some length, for a number of reasons, in nice ways, with good examples, with energy, with clarity and with experience. I was overruled.

What happened next was entirely predictable, and still somehow a joy to watch – not for the emotions the teams were put through, no, that was saddening, but the creative human spirit that emerges when people are forced into a game of measurement.

The teams were to be measured on velocity – roughly, how quickly a task travels from start to finish. And once people knew that was the number, they became remarkably, impressively and joyfully good at moving tasks from start to finish.

Small items like "Meet Dave" began appearing on boards and shifting again moments later – into the done state. Teams broke their work into ever tinier slivers. One team had genuinely mastered the form, with a run of tasks reading "Open document," "Edit document," "Save document," "Send document" – a masterpiece of the human capacity to royally outwit management.

The scoreboard was a marvel in real time. Every team became a thousand times more "productive." Nobody trailed for long; there was always another task to drag into Done. Most teams assigned one person to the job of working the board – their sole responsibility being to keep the team out of last place.

It was as clean a demonstration as I've ever seen of the oldest law in measurement: what you measure is what you get more of. The only casualty was the thing the whole exercise was meant to encourage. Real productivity – actual value, actually shipped – fell through the floor, because people were now spending more time playing the game than doing the work the game was pretending to reward.

In many of my workshops I run what I call the puzzle game. It's a child's game, somewhat ridiculous for a room of adults at work, and it teases out about twenty lessons that turn out to transfer directly to working life. It's also a lot of fun to play.

One of them is this: every team is working on different work. Some of it harder, some stranger, some more uncertain, some that changes shape while you're trying to pin it down and some that is more high pressured than others. And the teams themselves aren't copies of one another – they're made of people, and people vary, in engagement and skill and temperament and nerve and enthusiasm.

The work is different. The people are different. The context is different.

So why would you take a low-level number and compare it straight across all that variety – and, even with the best intentions in the world, why would you expect people not to game it the moment it's on a scoreboard?

I could have simply refused the charade. But it struck me, on the call, that the leader was reaching for a scoreboard because she was missing something every leader genuinely needs and rarely has: visibility and clarity. A way to see. The stick was a terrible answer to a real question.

So instead of arguing, I took a small team and we built something – a simple dashboard, run straight off the actual work-item data, with four levels for four different audiences. I'll describe it only as an example, not a template. The top level was financial: the line drawn between what was invested, the activity it is paying for, and the value coming back.

Below that, the quarterly commitments made against that investment – the outcomes and working product promised during planning. Below that, the granular work items, connected upward to those commitments. And at the bottom, closest to the ground, and probably the most important, the team's own numbers.

Each audience had its own altitude. Leadership lived in the top two – the money, and whether the commitments were trending with it. We invested this; we expected that; are we on track? The project teams used the middle. The teams used the lowest level: how good are we, really, at turning ideas into something real? Everyone could see every level, but each had their own view and their own use for it – and, crucially, the leaders could now leave the teams alone.

We trialled it, of course. A quick test. The leader took one look at the new visibility and stopped hunting for people to hit with sticks. The teams stopped gaming and started using their own numbers to get better. What we'd built was, in every way that mattered, a working portfolio system – not in the software, which is the part everyone wants to copy, but in the one thing that actually makes it work: a clean line drawn from investment, to the activity that investment paid for and only that activity, out to the value it was meant to create.

A team's numbers, in the team's own hands, are a mirror. They reflect the shape of the work back to the people doing it – where it flows, where it stalls, where the energy drains out of a week, where momentum returns after a hard one, where things can be improved or amplified. Held by the team, they help the team see itself, and therefore, help them to see how to get better.

The same numbers, in someone else's hands, become a lever (or a stick). A person far from the work reaches down, sets one team's throughput beside another's, and asks the question that sounds so reasonable: why are they faster than you? And the instant that question is asked, the mirror stops reflecting. The team begins performing for the audience instead of examining itself for the improvements and learnings.

Because a team's metrics only ever had one real job: to help that team ask where are we stuck, where are we repeating ourselves, where could this be smoother. They're instruments for looking inward, not scorecards for looking down. And they were never comparable in the first place. A number that every team seems to collect, usually from the tools they are using, fools us into thinking a team doing fifteen of something and a team doing eight are performing differently. But each of those is a team's own contextual measure, for its own contextual work, on its own contextual problems, for it's own contextual improvements. Comparing them is like comparing two novels by word count. Or two films by how long they run for.

A measure earns its place when it helps a particular person make a particular decision. At the team level, the only person whose decision the number can actually serve is the team itself. It's their mirror, and no one else's. The moment you turn it round to face the room, you ask the team to perform, not improve.

She never did get her stick. She got something better. And nobody added any tasks about meeting Dave.