On Making Things That Are Actually Good
"Good enough" is a project management concept, not a quality standard.
Somewhere in Christchurch, a few years ago, someone built a gate.
Not a grand gate — a side entrance to a property on an ordinary street. Timber, wrought iron, moderate scale. Nothing about it announced itself. But someone had built it with complete attention: the proportions held, the ironwork met the timber without apology, the whole thing swung and latched with the quiet confidence of something that knew exactly what it was for.
Nobody photographed it. Nobody wrote about it. It just stood there, doing its job, and doing it with a quality that most things don’t have — a quality you could feel before you could name it.
That quality has a name, actually. Good.
“Good enough” solves a scheduling problem, not a quality one.
The phrase enters projects during production, usually from someone with a deadline and a dependency. It means: this clears the bar we set, and we need to move on. A rational decision. A project management concept.
It does not mean the thing has the quality of the gate.
Conflating these two — treating “good enough” as a verdict on quality rather than a verdict on schedule — produces a particular kind of mediocrity: the mediocrity of things that function, that nobody complains about, that fill the space they were designed for, and that nobody would miss if they disappeared tomorrow. Most organisations generate this kind of work at scale. Not because the people lack talent. Because the question of goodness never entered the process.
“Done” and “good” optimise for different things. Done optimises for completion — for moving the dependency chain forward, for meeting the commitment, for demonstrating output. Good optimises for something harder to measure: whether the thing has earned its existence. Whether it does what it does with quality sufficient to justify its presence in the world.
Both matter. Most organisations only measure one.
Actually good doesn’t mean perfect.
This distinction collapses in both directions. “We can’t afford perfection” gets used to dismiss any standard above “good enough.” “We’re committed to quality” gets used to justify delays that have more to do with anxiety than with craft.
Neither position engages the real question.
Actually good means: fit for purpose, made with care, honest about what it attempts. A dish cloth, a wayfinding sign, a software onboarding flow — each of these can achieve actual goodness without approaching perfection. Goodness doesn’t require the best possible version of a thing. It requires enough attention to the thing to let it become what it should.
What goodness requires: someone took the problem seriously, made real decisions rather than defaulting to convention, and held something back from the world long enough to ask whether it had reached the quality that justified sending it out.
The gate in Christchurch wasn’t perfect. The timber would weather. The iron would need treatment eventually. But someone made it with the full weight of their competence, and that shows — in a way that doesn’t announce itself but doesn’t need to.
Organisations systematically produce the mediocre.
Not through laziness. Through structure.
Three forces push toward mediocrity in most organisations:
Absence of aesthetic authority. Nobody holds the right to say “this falls short” when the work technically meets its specification. Aesthetic judgements — about whether something feels right, sits well, carries its purpose with dignity — tend to get treated as optional, subjective, or obstructive. The people who hold them get overruled. After a while, they stop raising them.
Incentive structures that reward completion. Output gets measured. Quality — the kind that doesn’t show up in metrics — doesn’t. The person who ships ten adequate things and the person who ships five genuinely good things generate the same throughput number. Over time, this selects against the second person.
The cost of revision falls in the wrong place. Revising work takes time and pushes deadlines. The cost appears immediately and concretely. The cost of sending out something not quite good enough — the erosion of trust, the reader who puts it down, the object that gets ignored — appears slowly, diffusely, in ways that don’t trace back to the decision not to revise. So the revision doesn’t happen.
The result: not incompetent work, but competent work without distinction. Work that arrives and disappears. Work that doesn’t make anyone glad it exists.
The standard comes before the methods.
Most conversations about quality focus on process: better briefs, better review cycles, more time. These help. But they don’t address the prior question: does anyone here hold a standard?
A standard, in this sense, doesn’t mean a checklist or a rubric. It means someone — ideally several people — who carry a felt sense of what actually good looks like in this domain, who would recognise it if it arrived, and who would know with some certainty when the work hasn’t reached it yet.
Without that, process improvements produce faster mediocrity.
With it, even constrained processes can produce work of genuine quality — because the standard creates pressure that no methodology generates on its own. Someone looks at the work and says: not yet. Someone else asks what “not yet” means. A real conversation happens. The thing gets better.
The gate in Christchurch didn’t require unusual resources. It required someone who knew what a good gate looked like and cared enough to make one.
That person didn’t have a framework. They had a standard.
Having a standard at all.
The argument here doesn’t call for perfectionism. Perfectionism mistakes the pursuit of flawlessness for the pursuit of goodness, and produces work paralysed by its own standards. Actual goodness doesn’t have that quality. It moves. It ships. It accepts constraint. It lets the work become what it can become within real conditions.
But it starts from a different place than “good enough.” It starts from a genuine question: what would this need to become to justify its existence?
Not the best possible version. Not the version that wins awards. The version that earns its place — that does what it does with enough care and quality that, like the gate on the ordinary street in Christchurch, it gives whoever encounters it a small, quiet sense that the world contains things made well.
That sense accumulates. It shapes what people trust, what they return to, what they recommend without being asked.
“Good enough” never generates it. Only good does.