A change programme with nothing to show for six months is in trouble regardless of whether the underlying work is sound. Something visible has to happen early.
The advice is sound and it has a failure mode that is rarely stated, which is that the requirement for early results quietly becomes the criterion for what gets worked on.
What the early win is for
Its function is evidential rather than motivational.
An organization in month two has only the leader’s assertion that this will work. An early result replaces the assertion with an observation, and observations are what shift the prediction people are making about whether to invest effort.
It also supplies something specific to the people who committed first. They took a position on the leader’s word and are now exposed. A result vindicates them, which matters more than it vindicates you, because they are the ones who will carry the next stage.
What makes one count
Three properties, and most claimed early wins fail at least one.
Somebody outside the programme can verify it. A result attested only by the people running the change is not evidence, it is a report.
It is caused by the change rather than coincident with it. Attributing an ordinary fluctuation to the programme is transparent to anybody who knows the numbers, and it costs more credibility than the win supplies.
It matters to somebody other than the leader. A measure that improves and affects nobody’s day produces no belief.
The most useful early wins are frequently small removals rather than achievements: a duplicated form eliminated, an approval step removed, a report nobody read discontinued. They are verifiable, unambiguous, and they cost people nothing, which makes them credible in a way a claimed improvement is not.
The trap
The pressure for demonstrable progress starts selecting the work, and the selection happens without any decision being made.
Problems that produce measurable movement in a quarter get attention. Problems that would take two years to show anything do not, and those are frequently the ones the organization actually needed addressed. Nobody chooses to avoid the hard thing; it simply never wins the allocation against something that will report well in March.
The second form is worse. Where the visible measure diverges from the underlying objective, the measure wins. Response times improve while the quality of what is delivered declines, and the improvement is real and reported and the decline is not measured by anything.
The protection is to name the divergence risk at the outset. We are tracking this number because it moves early, and here is the thing it does not capture, which we will watch separately even though it is not going to look like anything for a year.
Claiming it without spending it
How an early result is presented determines whether it purchases anything.
Overclaimed, it converts into a liability. Announcing that a first result proves the approach invites everybody to test that claim, and the next unfavourable month becomes disproof.
The durable framing is narrow and forward-looking. Here is one thing that has changed. It is small and it is real. It is not the objective, and here is what we are doing next.
Credit the people who did it by name. The win’s most valuable function is showing that participation was rewarded, and that only works if the participants are visible.
Say what it is not
The most useful sentence attached to an early result is the one that prevents it being mistaken for completion.
Boards and funders are particularly prone to this: given a favourable indicator, they will treat the matter as handled and turn their attention elsewhere, which withdraws the pressure and the resources at the point where the difficult work begins.
This is early and it is fragile and the substantive problem is untouched, said at the moment you have good news, is uncomfortable and it protects the programme. A leader who reports only the favourable interpretation has borrowed against the second year.
When there is nothing to show
Some genuinely important changes produce nothing observable for a long time. Succession, culture, capability building.
Manufacturing an early win in those cases is worse than having none, because the manufactured version is recognised and it discredits the underlying effort.
The honest alternative is to report activity rather than outcome, explicitly labelled as such, with the timeline stated. Four people have completed the rotation. That is not a result and I am not going to present it as one. The result is whether we can fill the operations role internally in 2029.
That is a defensible position, it sets an expectation that can be held to, and it is considerably stronger than a fabricated metric that everybody discounts.
Edited by Patrick J. Wolf, PhD