Done > Perfect: Why people overdeliver and how leaders can break the cycle

Turning overcommitment into clarity, accountability and results.

Sep 21, 2025

Remember the movie Whiplash? A young jazz drummer drives himself to the edge of collapse, chasing flawless performance under a merciless teacher. It is an exaggerated Hollywood metaphor rather than a typical classroom, but it shows the trap clearly: when perfection becomes the goal, progress and even personal growth stops.

That reminded me of a phrase an ex colleague often used: “Progress over perfection.” At first it sounded reasonable. Who would argue against moving forward instead of polishing forever? But the more I thought about it the more it bothered me. Progress does not automatically equal value nor does it mean something is truly done. The phrase is vague, a bit like saying “Just make it better.” Progress can still leave big gaps: unresolved risks, unfinished compliance steps or ignored stakeholder needs. And in high stakes or legally regulated areas, mistaking “progress” for “completion” can even be dangerous.

On the other hand, perfection feels safe. But safety isn’t speed. And without speed even the best ideas stall.

This tension between the false safety of unnecessary perfection and the emptiness of meaningless progress is at the heart of this article. It explores a common leadership paradox: smart, motivated people who keep polishing, expanding or testing long after a task is “done.” It looks like commitment. It is often avoidance. In the following sections, we discuss why this topic is important and what research says about why this happens and how to help people escape this trap.

Why this matters

Perfectionism doesn’t just cost time. It blocks learning and slows innovation. Shipping creates feedback, alignment and real progress. Endless polishing creates delay. Whether you run a marketing team, a consulting project, a software program or an operations initiative, your role isn’t to make everyone “faster.” It’s to make them fast where speed is safe and flawless where perfection is truly required.

Leader’s lens: Treat perfectionism as a systemic issue (norms, incentives, risk framing) not a purely individual flaw.

What research tells us about perfectionism

Researchers consistently distinguish two faces of perfectionism:

Perfectionistic strivings1 → setting high goals, striving for excellence. Can be positive when paired with clarity and support.
Perfectionistic concerns2 → fear of mistakes, fear of judgment, self‑criticism. Drives stress, burnout, procrastination and unnecessary overwork.

Large scale reviews show: strivings correlate slightly with higher performance; concerns correlate with longer hours, stress and lower well being. In plain English: ambition helps, anxiety hurts.

Non linear reality: In complex work, more perfectionism isn’t always better. After a point returns can reverse as attention shifts from outcomes to impression management.

Why people overfulfill (across different contexts)

Think of overfulfillment as the cousin of procrastination dressed in a suit.

Fear of judgment: “If I add more nobody can say I didn’t try.”
Ambiguous finish line: Without a clear definition of “done”, people move the bar themselves.
Comfort in craft: Polishing feels safer than facing feedback.
Cultural norms: In “tight” contexts rough drafts look sloppy; in “looser” ones early sharing is pragmatic.
History and incentives: If early shipping was punished last time sanding the edges forever feels rational.
Remote/hybrid friction: Asynchronous comms slow feedback and increase “pre polish” to avoid misinterpretation.

Examples (general):

A marketing team asked for a 5 slide pitch deck delivers 40 slides of appendices. Decision fatigue, no decision.
A software team and product management delays release to “add just one more feature” missing real user feedback.
A consultant rewrites a report repeatedly while the client still waits for actionable recommendations.
A remote employee spends extra hours making a document “bulletproof” anticipating fewer chances to clarify later.

The psychology behind it

Perfectionistic concerns are tightly linked to procrastination: overpolishing delays evaluation. The mind trades the pain of possible critique for the comfort of extra effort. That comfort is expensive.

Spotting the pattern

Work expands late in the cycle without a new requirement.
Extra tests or sections appear without a risk based trigger.
Team members hesitate to show version 0.53 drafts.

The Mental Model: Two zones, one brain

Divide work into two zones:

Learning Zone (low risk) → Good enough, fast. Share v0.5 drafts, run quick pilots, expose assumptions early.
High Stakes Zone (high risk) → 100% compliance. Legal, safety, finance or regulatory deliverables. Perfection isn’t optional.

Calibration ritual (2 minutes)

Is this reversible? If yes → Learning Zone.
What’s the worst credible downside? If high → High Stakes Zone.
What is the cost of one more week? If high → ship sooner.

What to watch for (red flags of overfulfillment)

Gold plating4: Adding unasked for features or sections that delay delivery.
Silent polishing: Extra testing or editing without risk based rationale.
Scope creep via quality: Spec unchanged, internal bar quietly rises.
Volume as virtue: Far more material than requested.

Quantitative signals

Lead time rising while scope stays flat.
Review cycles per deliverable >2 by default.
After hours work up, outcome metrics flat.
Backlog aging (old items linger untouched).

How to lead people out of the “Perfection Trap”

Evidence informed moves you can start tomorrow:

1) Define “done” up-front

Use a one pager Definition of Done (DoD)5:

Purpose (one sentence)
Acceptance criterias (max 3)
Non goals (what’s out of scope)
Required artifacts
Stop rule (“We stop when …”)

You can find a One-Pager Template for a Definition of Done (DoD) with examples in the end of this article.

2) Separate speed from safety

Use two way6 vs. one way7 doors:

Two way (reversible) → bias for speed.
One way (irreversible/high cost) → slow down, apply full checklist.

3) Put risk on the table

Add a 3 box impact/probability screen to planning. Low impact → ship and learn. High impact (legal, safety, finance) → 100% compliance.

4) Install a gold plating kill switch

For extras, ask: Who asked? Which outcome moves now? What’s the cost of delay8? Vague answers → backlog.

5) Normalize version 0.5 (especially remote)

Schedule v0.5 reviews. Reward early sharing. Make rough work safe.

6) Embrace Iteration and MVPs

Adopt an iterative and incremental mindset: deliver in small slices that create value fast. An MVP9 (minimum viable product) is not a half baked product; it is the smallest useful version that tests assumptions and generates feedback. In marketing this might be a landing page before a full campaign; in consulting it might be a 2 page outline before a 50 page report; in product development it might be a basic feature shipped to a pilot customer. Each iteration should answer a question, reduce uncertainty and inform the next step.

7) Build “Psychological Safety”

Psychological safety is a team climate where people feel safe to speak up, admit errors and share rough ideas without punishment. This debriefs focus on learning, not blame. Script: What was the goal? What changed in the numbers? What did we learn? What’s next and what do we stop?

8) Teach Satisficing

Satisficing means choosing an option that is “good enough” to meet the need instead of chasing the absolute best. Set explicit thresholds tied to outcomes (adoption, cycle time, error rate). “Perfect” isn’t a metric; “<2% defects” is.

9) Limit “Work in Progress” (WIP)

Cap parallel work (WIP10). Shorter queues reduce waiting and the urge to polish endlessly.

Where leaders accidentally reward perfectionism

Praising volume over impact (“Impressive deck” vs. “This cut lead time by 12%”).
Ambiguous asks (“Can you polish this?” → infinite scope).
Slow feedback (encourages polishing while waiting).
One checklist for everything (prototype vs. final deliverable).
Leader modeling: Only showing perfect work teaches teams that rough drafts are unsafe.

Leader self check (quick audit)

Do I share v0.5 myself?
Do my asks specify length, scope, decision needed?
Do I green light extras without an owner or metric?

When perfection is non-negotiable

Perfection isn’t always the enemy. In certain domains it’s essential:

Healthcare and medicine: surgical procedures, drug safety.
Aviation and aerospace: flight critical systems.
Legal and finance: compliance and regulatory reporting.
Safety critical engineering: emergency systems.

The art is context sensitivity: knowing when “done is better than perfect” and when “perfect or nothing” applies.

Possible critiques of this view (and how to respond)

“You’re lowering standards.” → Not if you clearly mark High Stakes work. The point is differentiated standards, not lower ones.
“You’ll demotivate high achievers.” → Keep strivings high; reduce concerns by making feedback faster and safer. Ambition stays, anxiety drops.
“Culture eats process.” → True. That’s why leaders must translate expectations across tight/loose contexts and model v0.5 themselves.
“Metrics miss the craft.” → Use metrics to spot signals, not to replace judgment. Pair numbers with narrative reviews.

Getting started (without overthinking it)

This week

Define “done” visibly for one live task.
Label open items “two way” or “one way door.”
Add a stop rule to each deliverable.

This month

10 minute risk screen in planning.
WIP limits on the team board.
Run a premortem for one high stakes project.

This quarter

Different checklists for prototypes vs. final deliverables.
Ask managers to request v0.5 artifacts.
Track lead time, review cycles, cost of delay and discuss monthly.

Final thoughts

“Done” isn’t sloppy. It’s purposeful. Be fast where speed creates value. Be uncompromising where human safety, compliance or fiduciary duty are on the line. Teams don’t need more pressure. They need clearer bars and the safety to stop when they’ve reached them.

Think back to Whiplash: the student’s obsession with flawless performance delivered a dramatic finale, but also left behind exhaustion and fear. In business the real masterpiece is not endless polish but the rhythm of iteration, learning and knowing when to stop.

Bonus: Definition of Done (DoD) One-Pager Template (PDF)

Purpose (one sentence)
Acceptance criterias (max 3)
Non goals (what’s out of scope)
Required artifacts
Stop rule (“We stop when …”)

Definition Of Done One Pager Template V11

68.8KB ∙ PDF file

Download

Use it. Print it. Share it. That’s how behavior changes.