I found out exactly how bad my finishing problem was by accident. I ran a scan across my backups looking for something else entirely, and it surfaced 41 tasks that had been started and left mid-flight. Not 41 ideas. Forty-one things I had opened, touched, and walked away from.
This article is what those 41 items taught me, and the one rule I adopted afterward that changed my completion rate more than any productivity app ever did.
Why do I never finish what I start?
In my own case, quitting had little to do with willpower. I quit at the moment a project stopped needing more work and started needing something else — a decision, a reply, an approval, a reviewer. Nothing in my system noticed that stall, so the item silently rotted. When I read back through the 41, most of them weren’t abandoned. They were blocked.
That distinction matters, because the two problems have opposite fixes. If you’re lazy, you need discipline. If you’re blocked, discipline does nothing at all — you can grind for an hour on a task that is waiting on someone else’s email and end the hour exactly where you started.
The pattern shows up at scale too. Strava’s Year in Sport analysis, drawn from its global activity data, is where the popular idea of “Quitter’s Day” in the second week of January comes from — resolutions don’t fail in month six, they fail almost immediately. And research on gym memberships by economists Stefano DellaVigna and Ulrike Malmendier (American Economic Review, 2006) found that people on monthly contracts paid far more per actual visit than pay-per-visit pricing would have cost them. We don’t stop paying for our goals. We stop showing up for them.
What did a scan of 41 abandoned projects actually show?
It showed that abandonment has stages. The items weren’t scattered randomly across the difficulty curve — they clustered at specific handoff points, and almost none of them died because the work got too hard.
One caveat, stated plainly: the number I can stand behind is the total, 41. The stages below are how they sorted when I read through them afterward, not a tally I was keeping at the time. I’m not going to invent per-row counts to make the table look more scientific than it was.
| How far it got before stalling | What it looked like in the folder | The common stall point | Cheapest way to restart it |
|---|---|---|---|
| Named only | A folder and a title. Sometimes one note. | No first physical action was ever defined. | Write a short, concrete first action, or delete it today. |
| Researched, never started | Comparisons, saved links, options weighed. | Waiting to feel certain — an unmade decision wearing a research costume. | Decide in one line, using the information you already have. |
| Started and visibly working | A real draft, a running prototype. | Needed an answer from outside: a quote, an approval, a reply. | Send a one-line follow-up with a date attached. |
| Nearly done, never shipped | Complete but unpublished, unreviewed, unsent. | Nobody was waiting for it, so nothing forced the last stretch. | Give it an external audience and a deadline. |
| Done, never closed | Finished work that still read as “in progress.” | No closing ritual, so it kept occupying attention. | Mark it done. That’s the whole fix. |
Read the third column top to bottom and the theme is hard to miss. The stall almost never happened because the task was difficult. It happened at the seam — the moment the work needed something other than more work.
Where exactly do goals stall, and what is each one waiting on?
In my ledger, every stalled item turned out to be waiting on one of five things. I now keep that open-items ledger as the single source of truth for stalled work, and every entry is tagged by its blocker. That tag, not the priority field, is what determines whether an item ever moves again.
| Blocked on | What it feels like from the inside | Who can clear it | Real cost to clear | Clear it when |
|---|---|---|---|---|
| A decision | “I need to think about this more.” | You, alone, right now | One sentence | First. Always first. |
| Someone else’s action | “I’m waiting on them.” | A person you can actually reach | One message | Second — batch your nudges. |
| An external reply | “Still no answer.” | Outside your control | A follow-up plus a walk-away date | On a timer, not on your mind. |
| You (real work remains) | “I don’t have time for it.” | You | Hours or days | Only when it’s one of your two. |
| Time (not due yet) | Quiet. Nothing. | Nobody | Zero | Leave it parked and stop looking at it. |
The first row is the discovery worth stealing. Items blocked on a one-line decision are the cheapest progress available anywhere in your life, and they masquerade as the hardest ones. When the day is quiet, I clear those before anything else. A goal that has been dead for three weeks often needs a sentence, not a weekend.
Is this a motivation problem or a structure problem?
For me it was structure, and I suspect that’s more common than we admit. I spent 26 years inside one institution, and when I left I had to rebuild from nothing everything that place had quietly supplied: deadlines, review, and — the one I never noticed until it was gone — someone who noticed when a task stalled.
That third one is the invisible load-bearing wall. Inside an organization, an item that stops moving gets noticed by a colleague, a report, a meeting. On your own, a stalled item makes no sound whatsoever. It just stops.
Motivation research points the same direction. Peter Gollwitzer and Paschal Sheeran’s meta-analysis of 94 independent studies (Advances in Experimental Social Psychology, 2006) found that forming a specific if-then implementation intention produced a medium-to-large effect on goal attainment compared with holding the same goal without one. Same person, same desire, same goal — different structure, different outcome.
This is where the simple formula I work from earns its keep: Belief × Thinking × Action. It’s multiplication, not addition. Belief with no structure for action multiplies by zero. So does relentless action aimed at something you don’t actually believe you’ll finish.
How many things should you have started at once?
Two per project. After the scan turned up those 41 abandoned tasks, I capped work-in-progress at two started items per project. Ideas stay unlimited — capture everything, always. Only starting is rationed. When a third thing wants to begin, something has to close or be explicitly parked first.
This isn’t a mindset trick; it’s arithmetic. Little’s Law says average cycle time equals work-in-progress divided by throughput. Your throughput is roughly fixed. Double the things in flight and you double how long each one takes to finish — while feeling twice as busy the entire time.
| Unlimited starts | Two-started-items cap | |
|---|---|---|
| New idea arrives | Start it tonight | Log it; decide what it displaces |
| Cost of starting | Feels free | Visible and priced |
| What accumulates | Half-built things | A queue you can read |
| Emotional residue | Diffuse guilt with no address | Two clear obligations |
| Finishing rate | Falls as starts rise | Rises as flow narrows |
The hardest part is not the cap. It’s admitting that starting something new is a decision to delay something you already started.
Is abandoning a goal the same as failing?
No. Abandonment is maintenance. In my ledger, any item untouched for 30 days gets surfaced weekly as a discard candidate, and honestly, most of them get discarded. Deleting a goal you no longer want isn’t a loss — it’s the only way the list stays trustworthy enough to keep using.
Two well-documented biases fight you here. The Zeigarnik effect, from Bluma Zeigarnik’s 1927 work, describes how interrupted tasks stay lodged in memory more insistently than completed ones — which is why 41 dead items can weigh on you daily without a single one advancing. And the sunk cost fallacy, demonstrated by Hal Arkes and Catherine Blumer (Organizational Behavior and Human Decision Processes, 1985), keeps you funding projects purely because you already funded them.
A weekly discard sweep is the counterweight. It converts “I’m someone who never finishes anything” into “I finished this, and I decided against that.” Same facts. Completely different self-image — and self-image is the belief term in the formula.
Why can’t you find your own past notes when you need them?
Because you filed them by topic instead of by symptom. A personal knowledge system only compounds when each index line describes the problem you were having, not the subject you were writing about. You search at the moment of pain, using the words of the pain — and topic-shaped labels never match those words.
The difference is small and total:
- Topic line: “Notes on deployment configuration” — findable only if you already remember it exists.
- Symptom line: “Build passes locally, fails on the server — check environment variables first” — findable by anyone standing in that exact hole, including you, two years later.
This is how I built and shipped working software without being able to read code: I wrote down every point of failure the moment it happened, described as the failure rather than as the feature. The record of what broke became the manual for what to do next time.
How do you start finishing this week?
Do the audit before you do anything ambitious. Scan your folders and lists for everything you have actually started. Write the total down — the number itself does most of the persuading. Then tag each item with what it’s waiting on, clear every one-line decision today, pick two to keep, and let the 30-day sweep take the rest.
What you’ll find is that you were never a person who couldn’t finish. You were a person running an institution’s worth of work without any of the institution’s machinery for noticing a stall.
If you’d rather have that machinery running for you than maintain it by hand, the VisionDream app is built around the same idea — a vision and goal app that breaks long-term goals down into daily habits you actually track. And if you want the longer version of how I rebuilt this after leaving a 26-year career, it’s in my book, ‘Why Do I Fail While Others Succeed?’
Forty-one abandoned projects didn’t tell me I lacked discipline. They told me exactly where the seams were. Once I could see the seams, finishing stopped being a character trait and became a setting I could change.