Performance Management in a Changing Landscape – Redefining Success
Redefining Success: Why Performance Is a Proven Architecture, Not a Posted Result
Every conversation about performance management skips the question it depends on: what is success? We argue about how to measure it, how often to review it, what to reward, and rarely define the purpose that all of that is supposedly in service of.
And the definition matters more than any of the mechanics, because if you get it wrong, every mechanic built on top of it is calibrated to the wrong target.
There are two conventional answers, and both are traps.
The first answer is that success is hitting the targets. The metrics met, the KPIs are green. This fails for a reason worth stating precisely: the moment a measure becomes the definition of success, people optimise the measure rather than the thing it was standing in for. The proxy displaces the purpose. An organisation can hit every target and still fail at what the targets were meant to represent, because the targets became the game.
The second answer looks safer and is more insidious: success is the result. The revenue, the delivery, the outcome achieved. The problem is that the result is a lagging output, and it gets attributed to the people, when the result was actually produced by the architecture they worked inside.
Define success as the result, and you’re back to praising the person who happened to stand at a favourable point in the structure and blaming the one who stood at an unfavourable one. The exact person-not-structure confusion that misreads everything. Worse, two organisations can post identical results while one produced it reproducibly, by design, and the other extracted it heroically, from specific exhausted people who will burn out or leave.
A result-definition of success judges the two as equal. They are not. The second didn’t succeed; it borrowed against a structure it isn’t sustaining, and the result is masking a failure already in progress.
Success is an architecture proven, not a result achieved
So success can’t live at the level of the metric, and it can’t live at the level of the result.
It has to live at the level you can actually build: the architecture. An organisation has succeeded when it has built a structure that reliably turns the capability of ordinary competent people into genuine value for the people the work serves, reproducibly, without depending on heroics, and without the proxy displacing the purpose.
Unpack that, because each part is load-bearing.
Genuine value to the people the work serves: the stakeholder outcome, not the metric that stands in for it, is what the architecture is actually for; naming it this way is what stops the measure from becoming the mission.
Reliably and reproducibly, as the rational output of the structure, this is the difference between a structure that produces performance by design and a lucky or extracted result that won’t repeat.
And without heroics, because a result that depends on someone carrying an unsustainable load isn’t success, it’s deferred failure. The architecture is consuming exactly what it needs to keep going, and the dashboard won’t show it until the person breaks.
You have succeeded, in other words, when good performance has become the rational, repeatable output of the structure you built, not the lucky or extracted output of the people you happened to have. The result is how you read whether the architecture is sound. It is not the thing you chase.
Why this makes performance management a development function, not a measurement one
Once success is defined as a proven architecture rather than a posted result, what performance management is for changes completely.
If success were the number, the job would merely be a measurement.
Because success is a structure that reliably converts capability into stakeholder value, the job is to build that structure and to develop the capability it converts.
Performance management is a development function.
Measuring output is, at most, a way of checking whether the architecture is working, a readout, never the point.
This is why the traditional metrics-first model fails now, and the failures are specific.
It biases toward the short term, because a target is a snapshot of what mattered when it was set, and defending it crowds out the slower work of building capability. It flattens people, because a standard metric captures a narrow band of contribution and renders the rest invisible.
It resists adaptation. Because someone measured against a fixed target is structurally discouraged from the pivot new conditions require, since pivoting reads as missing the number. And it measures outputs while ignoring the capability that produced them, rewarding this quarter’s result while doing nothing to build next quarter’s. None of this is an argument against measurement as an instrument.
It’s an argument against mistaking the instrument for the definition of success, and mistaking measuring for managing.
What capability actually means
If the job is developing capability, capability has to be defined as specifically as success was; otherwise, “develop your people” is just another slogan. It isn’t vague. Drawn from what the work actually demands, it’s the competence to keep a process reliable rather than firefighting it, to strip unnecessary complexity out of how the work is done so it becomes more dependable, and to improve and innovate at the point where the work meets the people it serves.
Developing performance means developing that, the real, definable competence the role requires, because that competence is precisely the input the architecture converts into stakeholder value.
Adaptability is built, not hired or exhorted
The capability that matters most in a fast-moving environment is adaptability:
The ability to learn, unlearn, and adjust as conditions change.
But adaptability is where the exhortation reflex is strongest and weakest.
Telling people to be more adaptable, or hiring for it as a trait, achieves little because adaptability is far more a property of the structure than of the individual.
Put a genuinely adaptable person in an architecture that punishes the failed experiment, routes every change through six approvals, and measures them against a fixed target, and they stop adapting within a quarter, because the structure rewards standing still.
Put an ordinary team in an architecture where learning is safe, where they hold the decision rights to adjust their own work, and where improving the method is rewarded rather than treated as deviation, and they adapt.
You don’t cultivate adaptability as a mindset in individuals. You build the architecture that makes adapting rational, and then it becomes a property of everyone inside it.
Where this leaves the ritual, and the manager
Two things follow, both developed at length in their own right, so the point here is only how they connect to the definition.
The formal annual assessment has to go, not because feedback is bad, but because a scheduled, backward-looking verdict is the wrong instrument for developing capability. Evaluation has to be embedded in the work, continuous, and owned by the people closest to it, functioning as real-time input to growth rather than a compliance exercise. And competency development has to be specific, targeted at the actual competences a role demands and how the person is growing in them, rather than a uniform training catalogue.
This redefines the manager.
In the measurement model, the manager conducts the review.
In the development model, the manager builds and holds the architecture within which capability develops, and stakeholder value gets produced: the clarity about what good work is, the information flows that make performance visible in real time, the decision rights that let people improve their own work, and the safety that makes learning from failure possible rather than career-ending.
And beneath that architecture sits the foundation it expresses.
Performance, behaviour, and the capacity to adapt are all produced by the structure – the decision rights, incentives, and consequence pathways – and that structure is itself the expression of the organisation’s culture: its real answer to what actually gets rewarded here.
Culture is the foundation; the architecture built on it is what produces performance. Behaviour and adaptability are the outputs.
That’s why performance can’t be fixed at the level of process, and why it isn’t a matter of “stewarding culture” as though culture were a garden to tend.
You improve performance by building the architecture that develops and rewards capability, not by measuring output harder, and not by exhorting people to grow.
The point
The whole apparatus of performance management was built to answer the wrong question: How did people score against the targets? When the question that actually decides an organisation’s future is whether it has built a structure that turns ordinary capability into genuine value, reliably and without burning through its people to do it.
That’s what success is: not a result you posted, but an architecture you proved.
Stop asking how to measure performance more precisely.
Ask what your structure is actually built to produce, whether it can produce it again next year without heroics, and whether the value “reaching your stakeholders” is real or just a proxy for it looking healthy on a dashboard.