Performance management starts before the dashboard

Flowzi is meant to become the engine of a migration agency. If it runs the work, then a leader will — rightly — expect it to tell them how the team is doing. Who’s carrying what. Where the bottlenecks are. Who’s stretched and who has room. What next quarter’s capacity actually looks like before they sign the next big client.

That expectation is fair, and we’ve built for it. Performance tracking and a real read on how staff and clients behave aren’t a bolt-on report; they’re in the core of the system. If Pulse is the engine, the instrumentation comes with the engine.

But I want to write down the part that the dashboard can’t tell you, because it’s the part I’ve watched people get wrong — including me.

Performance management is not a data problem. Collecting the numbers and presenting them well is the easy 20%. The hard 80% is human, and most of it happens before a single metric is ever recorded.

It starts at recruitment

The cheapest hour you’ll ever spend on performance management is the hour you spend hiring the right person.

A team that invests real, unhurried time getting the fit right — the skill, yes, but also the temperament and the way someone works — saves itself countless hours of correction, coaching, and difficult conversations later. Hiring well isn’t a cost centre that competes with “the real work.” It is the real work, done early. The best performance intervention on any file is the one that already happened, quietly, in the interview room months before.

No dashboard rescues a bad hire. It just gives you an earlier, more precise view of a problem you created upstream.

A dip is a signal, not a verdict

Here’s the trap the instrumentation makes easy to fall into.

You hire someone after several rounds. They’re good — you’ve seen it. Then, months in, the numbers soften. Cases sit longer. Follow-ups slip. The chart turns the wrong way.

The lazy read is that you misjudged them, or that they’ve stopped caring. Almost always, neither is true. People you carefully chose don’t suddenly become different people. What’s usually happening is life: an illness, a parent who’s unwell, a relationship coming apart, a financial shock, something at home they haven’t told you and may never. The dip is real. The story the dashboard implies about why is almost certainly wrong.

So the number’s job is narrow and honest: it tells you where to look. It does not tell you what you’re looking at. A leader who treats the dip as a verdict loses good people who were three weeks away from being fine. A leader who treats it as a signal walks over and asks a human question.

Motivation isn’t one thing

The other reason the data can’t manage anyone is that people aren’t running on the same fuel.

For one person, the driver is career progression — a title, a harder brief, a path they can see. For another it’s the bonus, plainly and without apology. For a third it’s recognition, or autonomy, or simply being trusted with the messy cases nobody else wants. Pay the second person in praise and the first person in cash and you’ll frustrate both while thinking you’re being generous.

Finding what actually moves each person is the leader’s job, and there’s no shortcut and no instrument for it. It comes out of one-on-ones — the regular, slightly boring, genuinely-listening kind — long before you need it. You want to already know what a person is chasing on the ordinary Tuesday, so that when their numbers dip you have the relationship to ask why, and the map to help them back.

So what is the system actually for?

Not to manage people. To help a leader manage people well.

Used honestly, it does a few things a human can’t do alone:

• It’s a leading indicator, not a post-mortem. A good read surfaces the softening early — early enough for the caring conversation, not the exit interview. The point of seeing the dip in week two is to help in week two.
• It makes fairness possible. Consistent visibility across the whole team is a check on gut-feel and favouritism. You praise and promote on what’s real, not on who sits nearest you or speaks up loudest.
• It protects people from overload. Workload and capacity you can actually see let you distribute the work before someone burns out quietly and resigns loudly. That’s planning and care in the same view.

And one thing it must never become: a stick. The moment a metric becomes the target, people optimise the metric instead of the work — and the number stops meaning anything. Measurement changes behaviour; that’s exactly why you manage the person and not the chart. Data measures. Leaders manage. The system’s job is to hand a good leader a sharper, fairer, earlier view — and then get out of the way of the human conversation that does the actual work.

It’s the same conviction that runs through the rest of what we build: the system surfaces the signal; a person makes the call. Suggested, not dictated — for clients, and for the team that serves them.

Build the engine well enough and it’ll tell a leader exactly where to look. What they do when they get there is still, thankfully, a human being’s job.