The "Done" Definition That Saves Time

Welcome to The Logical Box!

For leaders who want AI to help, not add more work.

Hey,

If this is your first time here, welcome! The Logical Box is a weekly newsletter for owners and leaders who want AI to reduce real work, not add new work. Each issue focuses on one idea: see where work breaks down, fix the clarity first, then add AI where it actually helps.

If your business only works because you are in the middle of everything, this newsletter helps you build systems, so it does not have to.

Last week I showed you what happens when unclear ownership meets AI. Speed without ownership creates clutter that moves faster.

This week I want to show you the other half of that problem: what happens when "done" is never defined.

Because even when one person owns the work, if they do not know what "done" looks like, the work cycles forever.

The proposal that never closed

A consulting team spends three weeks on a proposal. The owner writes the first draft. Good start. Clear structure. Solid content.

Then it goes to review.

One person adds detail to the pricing section. Another person rewrites the intro to sound "more strategic." A third person moves two sections around because they think it flows better.

Each change is reasonable. Each reviewer is trying to help.

But nobody ever agreed on what "done" meant. So the proposal keeps getting rewritten. Not because it was wrong. Because nobody could point to a clear standard and say "we hit it, ship it."

The client was ready to move forward after week one. The team spent two more weeks revising a document that was already good enough to win the work.

That is not perfectionism. That is the cost of starting without a finish line.

AI does not help when "done" is fuzzy

Now add AI to that same scenario.

The owner uses AI to write the first draft. Two hours instead of two days. Great. Real time saved.

Then the same cycle starts. One person feeds the draft back into AI and asks for "a more strategic tone." Another person runs it through a different tool to "tighten the language." Someone else regenerates the pricing section because they think the format could be clearer.

Now you have four versions. All slightly different. All reasonable. And nobody knows which one to send because "done" was never defined before the first draft was written.

AI did not create the problem. It just made the loop faster. Same confusion. Same wasted time. Just with more polished outputs sitting in a folder.

Remember last week's data from Workday and Hanover Research: nearly 40% of AI time savings are lost to rework. This is where that rework happens. Not because the AI output was wrong. Because the team never agreed on what right looks like.

What "done" actually means

Here is what most teams miss. "Done" is not about perfection. It is about agreement.

Done means: we can point to this output, compare it to a simple standard, and decide whether it meets the need.

That standard does not have to be complicated. It can be one sentence.

For a client proposal: "Done means pricing is clear, scope matches what we discussed, and the next step is obvious."

For a weekly update: "Done means the three priorities are listed, blockers are named, and nothing is vague."

For a customer onboarding email: "Done means the new customer knows what happens next and who to contact if they have questions."

You are not writing a rubric. You are drawing a line so the team knows when to stop revising and start moving.

When "done" is clear, AI becomes useful

Let me show you what changes when you define "done" before AI touches the work.

Same consulting team. Same proposal task. But this time the owner starts with a two minute conversation.

"Done means: three service options with clear pricing, one case study that matches their industry, and a simple next step they can take this week. If it has those three things and reads clearly, we send it."

Now the owner asks AI to draft the proposal with that standard in mind. AI delivers a solid first version in two hours.

The owner checks it against the three criteria. Pricing is clear. Case study matches. Next step is obvious. Done.

One reviewer glances at it to confirm. No rewrite. No second guessing. The proposal goes out the same day.

The client responds within 48 hours. Work moves.

Same AI. Same tools. Different outcome because "done" was named before the work started.

That is how AI creates value instead of confusion. Not by producing better drafts. By producing drafts that can be finished because the finish line was drawn first.

The test: can you say it in one sentence?

Before your team starts any task this week, try this.

Ask: "What does done look like for this?"

If the answer takes more than one sentence, the task is not clear enough yet. Simplify it.

If nobody can answer, stop. Do not write the draft. Do not open AI. Have a two minute conversation to define the finish line first.

This is not extra process. This is the thing that prevents the process from spiraling.

When "done" is clear, the owner knows when to stop. The reviewer knows what to check. AI knows what to optimize for. And the work moves instead of cycling.

Where this breaks most often

I see three places where "done" goes undefined and costs the most time.

Collaborative documents. Three people editing one file. Nobody knows whose judgment wins. The doc gets revised until someone finally says "we are out of time, just send it."

Recurring tasks. Weekly reports, client updates, status emails. The format drifts every week because "done" was never written down. So every week starts from scratch.

AI assisted drafts. The tool creates output fast. But if "done" is not clear, people keep regenerating and tweaking. The two minutes you saved disappears into thirty minutes of indecision.

If any of those sound familiar, the fix is the same. Define done before you start. One sentence. One standard. Then let the work hit that line and move.

What to protect

Defining "done" does not mean lowering standards. It means naming them.

You still get to be precise about what matters. You still get to care about quality. You just stop pretending the standard lives only in your head.

And here is the part nobody talks about: when "done" is undefined, you stay in the middle of everything. The team routes decisions back to you because they do not know when to stop working.

When "done" is clear and written down, the team can finish without you. That is not about delegation. That is about freedom.

AI can help you work faster. But if you do not define "done," you just produce more drafts that sit in limbo while everyone waits for someone to decide.

Name the line. Then let the work cross it.

Try this

Pick one task your team will do this week. Before anyone starts, write one sentence that describes what "done" looks like.

Not perfect. Not ideal. Just done.

Then let that sentence guide the work. When the output matches the sentence, ship it.

You will finish faster. You will revise less. And your team will stop waiting for you to tell them when to move.

Build your skills here!

I launched AI Clarity Hub, a private space for owners and leaders who want AI to reduce real work, not add new work.

Inside the hub, we work through exactly what I described today: finding the clarity gaps, building simple systems, and only then adding AI where it actually helps. Members get access to live training sessions, ready to use templates, and a library of AI assistants built for real business workflows.

Thanks for reading,

Andrew Keener
Founder of Keen Alliance & Your Guide at The Logical Box