The project is almost done. The design is finished, the draft is written, the code mostly works. What is left is small: a few fixes, one email, a final read-through, the part where you press send. And it has been sitting there for three weeks. If ADHD and finishing projects feel like two things that refuse to meet, you are not imagining it. The last stretch of a project is a different kind of work from the rest of it, and it happens to be the kind of work ADHD brains find hardest to start.
This article is about that last stretch specifically. Not why you never start (that is a separate problem with its own fixes), and not the general pattern of leaving tasks unfinished. It is about the project that is 90 percent done and will not move, why it stalls exactly there, and how to get it over the line.
The 90 percent stall is a real pattern
Software engineers have joked about it for decades. The "ninety-ninety rule," credited to Tom Cargill at Bell Labs, says the first 90 percent of a project takes 90 percent of the time and the last 10 percent takes the other 90 percent. It is a joke because it is true for almost everyone. The finishing work is underestimated, scattered, and boring.
For ADHD brains the effect is sharper, and the reason is not effort. Most people with ADHD who stall at the end of a project have already done the hard part. They solved the interesting problem, built the thing, wrote the chapter. The stall is not a failure of ability. It is a mismatch between what the last stretch asks for and what an ADHD brain is wired to supply.
What the last 10 percent actually asks of you
Look closely at what is usually left when a project stalls near the end. It tends to fall into four kinds of work, and none of them look like the work that came before.
Small, scattered tasks. The middle of a project is one big thing you can sink into. The end is twelve small things: fix the typo on page four, export the file at the right size, check the one link, rename the folder. Each one is trivial. Together they demand constant switching, which carries a real cognitive cost and gives none of the absorbing focus that carried you through the middle.
Work with no novelty left. By the end you know everything about the project. There is nothing left to discover. ADHD is closely tied to differences in how the brain's reward system responds to novelty and interest, and research on dopamine signalling in adults with ADHD (Volkow and colleagues, 2009) points to a reward pathway that responds less to tasks that are not inherently engaging. The beginning of a project is full of novelty. The end has none.
Exposure. Finishing means showing it to someone. Sending the report, publishing the post, handing in the thesis. Unfinished work is safe because it cannot be judged yet. Many people with ADHD carry years of criticism for missed details, and the moment of release is exactly where that history shows up as a vague, heavy reluctance.
An undefined finish line. "Done" is rarely written down. Without a clear definition, the end keeps moving: one more pass, one more improvement, one more check. The project never stalls at a decision point. It drifts.
Why the finish line does not pull ADHD brains the way it should
Psychologists have a name for the extra push most people feel near a goal: the goal-gradient effect. First described by Clark Hull in 1932 with rats running faster as they approached food, it was later shown in people by Ran Kivetz, Oleg Urminsky, and Yuhuang Zheng in 2006, who found coffee-card customers bought coffee more often as they got closer to a free drink. The nearer the goal, the harder people work.
The goal-gradient effect depends on the finish feeling close and rewarding. For an ADHD brain at 90 percent, neither is guaranteed. The reward for finishing is often delayed (the grade comes later, the client replies next week), and delayed rewards are discounted more steeply in ADHD. The finish line also feels far, because the remaining work is fuzzy and spread out. So the pull most people feel at the end is weak, right when the work itself has become least interesting.
There is a second force working the other way. Unfinished work stays active in memory, a pattern known as the Zeigarnik effect. The almost-done project keeps surfacing at 11 p.m., in the shower, in the middle of something else, each time with a small spike of guilt. That mental noise makes the project feel heavier than it is, which makes starting the last stretch harder still.
How to cross the line
The fixes that work for the last 10 percent are different from the ones that help you start or keep going. They are about making the end concrete, small, and short.
1. Write the finish line down
Before you do anything else, define done in one sentence. "Done means the PDF is emailed to the client." "Done means the post is live and shared once." Not "done means it is good." A written finish line stops the end from drifting and turns a vague feeling of unfinishedness into a target you can actually hit.
2. List every remaining piece, including the tiny ones
Most stalled projects are blocked by work nobody has written down. Spend ten minutes listing every remaining step, down to "find the client's email address." This is the same move as a brain dump, aimed at one project. Seeing that what is left is eleven small items, not an endless fog, shrinks the project in your head. It also tells you which piece is genuinely blocking the rest.
3. Do the smallest piece first, on purpose
Pick the item that takes under five minutes and do it right now. Not the most important one, the smallest one. Movement on a stalled project creates its own momentum, and lowering the activation energy of the first step is the single most reliable way to restart work an ADHD brain has stopped seeing.
4. Batch the scattered work into one sitting
The end of a project is made of fragments, and switching between them all day costs more than the fragments themselves. Put the small tasks into one time block, ideally 45 to 90 minutes, and do them back to back. A timer helps. So does doing it next to someone, in person or on a call, which is why body doubling works so well for exactly this kind of dull, finishing work.
5. Separate finishing from polishing
Most 90-percent projects are not unfinished. They are unpolished, and the polishing has no end. Decide what is "good enough to ship" and finish to that, not to perfect. If you genuinely care about polish, schedule it as a separate version two. Shipping version one is the goal.
6. Make the release small and immediate
The moment of exposure is easier when it is a small, fast action rather than a big event. Draft the email now, attach the file, and schedule it, or ask one trusted person to look at it before you send it widely. Lowering the stakes of the final step often removes the reluctance that was holding the whole project in place.
7. Use a deadline that someone else can see
External pressure is the one force that reliably drags ADHD work across the finish line. If the project has no real deadline, borrow one: tell a friend the date, book a review meeting, or commit publicly. A deadline that exists only in your head is easy to renegotiate. One that someone else is waiting on is not.
When to let a project go instead
Not every stalled project deserves a finish. Sometimes the 90 percent stall is information: the project mattered when you started and does not any more. The guide on how to finish what you start covers this decision in more depth, but a quick test helps. If you imagine the project finished tomorrow and feel relief, finish it. If you feel nothing, or feel trapped, it may be time to close it on purpose, with a sentence written down about why. Deliberately ending a project is not the same as abandoning it, and it clears the mental noise just as well as finishing does.
A short checklist for the next stalled project
- Write down what "done" means in one sentence.
- List every remaining step, including the tiny ones.
- Do the smallest step right now.
- Put the scattered work into one time block, ideally with a person nearby.
- Ship good enough, and schedule polish separately.
- Make the final release a small, fast action.
- Give the deadline to someone else.
The real problem was never effort
If you have a folder full of projects that stopped at 90 percent, it is easy to read that as proof that you cannot finish anything. The pattern says something more specific. You can do the hard, interesting part. What stalls you is the boring, scattered, exposed last stretch, at the exact moment the reward feels furthest away. That stretch can be made smaller, clearer, and shorter. Most stalled projects need one focused hour, not more discipline.
FAQ
Why do people with ADHD struggle to finish projects?
The last part of a project is usually small, scattered, and free of novelty, while the reward for finishing is often delayed. ADHD brains respond less strongly to tasks that are not inherently interesting and discount delayed rewards more steeply, so the end of a project gets the least drive at the moment it needs the most.
Is it normal to stop working on a project when it is almost done?
Yes. The "ninety-ninety rule" exists because the last 10 percent of most projects is underestimated by almost everyone. ADHD makes the stall more common and more pronounced, but the pattern itself is universal.
How do I finish a project when I have lost interest?
Define what done means in one sentence, list every remaining step, and do the smallest one immediately. Batch the rest into one short session and ship it at "good enough" rather than perfect.
Should I finish every project I start?
No. Some projects stop mattering. If imagining the project finished brings relief, finish it. If it brings nothing, close it deliberately and write down why, which frees the same mental space as finishing.
Originally published at https://www.habidu.com/news/adhd-finishing-projects
Top comments (0)