Finishing What You Start: 5 Failure Points and Checkpoints
Key takeaways
- Projects abandon at five predictable points: the novelty spike, the boring middle, ambiguity, the first setback, and the 90% stall.
- Set a written checkpoint at each failure point in advance, because willpower cannot fix any of them.
- Run a deliberate-quit test in writing: two noes out of three questions means archive the project with a dated decision.
- Restart an abandoned project with a 20-minute archaeology session followed by one small quick win, with no fixing allowed in the session.
- Gamified tools help most when they reward the boring middle, but the checkpoint you write down matters more than any XP curve.
You don’t finish what you start because projects die at five predictable points: the novelty spike, mid-project boredom, ambiguity, setback, and the near-done stall. Willpower can’t fix any of them. A specific checkpoint at each one can, and this piece gives you the question, trigger, and pass/fail test for every point.
Why do I keep not finishing what I start?
You don’t abandon projects at random. You abandon them at the same handful of points, every time, which is why the app graveyard on your phone keeps growing even though each new tool felt like the fix.
The pattern is predictable, not random
Projects die between start and done, not at either end. The novelty spike carries you through week one, then boredom, ambiguity, setbacks, and the near-done stall each claim their victims. If you can name the exact point where you bailed last time, you can set a checkpoint there next time.
Why ADHD brains abandon at the same five points
ADHD affects task initiation and sustained follow-through, not effort or intent (CHADD). Interest and urgency drive attention, so once a project stops being new, the dopamine stops and the project stalls. That’s why gamified XP and streaks help: they manufacture fresh triggers mid-project. SideQuest’s Boss Battles exist for exactly this.
What are the five points where most projects die?
Projects don’t die randomly. They die at five predictable points, and knowing the map is half the battle.
Point 1: the novelty spike fades (days 3-10)
The first week runs on dopamine. New app, new plan, hero moment energy. Then the spike drops and the project is just work. Most app-hopping happens here, not because the tool failed but because the novelty did.
Point 2: the boring middle
No XP drops feel big anymore. The middle is where streaks break, because nothing about it feels like a quest.
Point 3: ambiguity about the next step
“I’ll work on it” is not a next step. When you can’t name the exact next action in one sentence, your brain picks literally anything else.
Point 4: the first real setback
One bad day, one bug, one rejected draft. The setback reads as evidence you’re not the kind of person who finishes.
Point 5: the 90% stall
Nearly done projects stall because finishing has no novelty left and no clear reward attached.
| # | Failure point | Typical timing | Feels like |
|---|---|---|---|
| 1 | Novelty spike fades | Days 3-10 | “This app is boring now” |
| 2 | Boring middle | Weeks 2-6 | “I’ll come back to it” |
| 3 | Ambiguity | Any time | “I don’t know what’s next” |
| 4 | First setback | Unpredictable | “Maybe this was a bad idea” |
| 5 | 90% stall | Near the end | “It’s basically done anyway” |
The fix isn’t willpower at each point; it’s a checkpoint you set in advance. That’s what the next section covers, and it’s the same logic we use in our own builds (about us).
What checkpoint should I set at each failure point?
Checkpoint 1: the day-7 honest question
Novelty carries you for about a week, then the dopamine drops and the project competes with the next shiny thing. On day 7, ask one question out loud: “Would I start this today if I hadn’t already?” If the answer is no, either shrink the project or kill it now, before you sink another month. Day 7 is cheap; day 30 is a graveyard.
Checkpoint 2: shrink the quest, not the goal
Mid-project boredom usually means the next step is too big to feel like a quick win. Keep the goal, cut the quest size: not “write the report” but “write the ugly first draft of section two.” If a step takes more than one sitting, it’s two quests. Gamified task apps work on exactly this principle, and the feature checklist that matters for ADHD brains is spelled out in task management features for ADHD.
Checkpoint 3: define ‘next action’ or quit
Ambiguity kills quietly. If you cannot name the very next physical action without thinking for more than ten seconds, the project is stalled at ambiguity. Write the next action or formally pause the project. “Figure out the app” is not a next action. “Install the SDK and run the starter template” is.
Checkpoint 4: the setback triage rule
When something breaks, you get 24 hours to triage: fix, route around, or descope. No re-planning the whole project from scratch, which is how setbacks turn into abandonments. One miss is data. Three consecutive misses means the quest is mis-sized, and you drop back to checkpoint 2.
Checkpoint 5: the finish-line audit
Near-done stall is real: unfinished tasks nag at you even when you’re doing something else, which psychologists call the Zeigarnik effect, where incomplete tasks stay more memorable than finished ones (Wikipedia, Zeigarnik effect). That open-loop tension can work for you if you audit what “done” actually requires. List every remaining step, mark which ones are load-bearing versus nice-to-have, and cut the nice-to-haves in writing. A project that ships at 85% beats one that dies at 95% with a polish backlog you’ll never open. The audit takes fifteen minutes and converts a vague guilt cloud into a short, finite list you can actually defeat like a boss battle.
The honest caveat: these checkpoints are rules you have to remember to run, and no checklist survives a bad week. Set them as calendar repeats, not intentions, because a checkpoint that lives in your head is just another open loop.
How does the importance of finishing what you start compound over time?
Finished projects are the only real XP
Starting feels like leveling up, but it awards nothing. XP only banks when a quest is marked done, and every abandoned project is a character with zero progress saved. Finishing is also a skill with reps: each completion teaches your brain that the boring middle and the near-done stall are survivable, which lowers the activation energy for the next one.
Why one completion beats five abandoned starts
Five fresh starts give you five novelty spikes and five crashes. One finished project gives you momentum, proof, and a template you can reuse. That gap compounds: finishers get faster at finishing, while starters get better at starting and nothing else.
The fix is small: pick one stalled project and cut it down until it’s a quick win, then bank it. If task paralysis is what froze it, name the type first. One hero moment per week beats five downloads per month.
How do I decide whether to finish or deliberately quit a project?
The deliberate-quit test
Answer three questions in writing, not in your head: Does this project still solve a problem I actually have? Would I restart it today if I found it fresh? Is the remaining work specific enough to schedule? If two of three are no, quit on purpose. Write a one-line decision and a date, so the quit becomes a completed task instead of an open loop draining XP from everything else.
Quitting on purpose vs. quitting by drift
Drift is what happens when novelty spikes fade and no checkpoint forces a decision. Deliberate quitting is an if-then plan: “If I skip this project for 14 days, I will either schedule the next quest or archive it.” Gollwitzer’s research on implementation intentions shows that pre-deciding your response to a trigger dramatically increases follow-through, compared with vague intentions. The same structure works for restarts, and habit tracker ideas like a “quit log” or a deliberate-archive streak turn the decision itself into a quick win.
Honest caveat: a gamified tool can make quitting feel like losing progress, which nudges you to keep zombie projects alive. Use the written test, not the streak, to decide.
How do I restart a project I abandoned months ago?
The 20-minute archaeology session
Set a timer for 20 minutes and do nothing but recover state. Find the last file, note, or commit. Write three lines: where the project actually stopped, what the next physical action is, and what confused you when you quit. That last line matters most, because ambiguity is usually what stalled you, not laziness.
Do not fix anything in this session. You are a detective, not a builder.
Re-entry without shame
Restart at the smallest possible quest, not the point where you left off. If you stalled mid-project, your first task back is a quick win: rename a file, sketch one screen, write one paragraph. The goal is momentum, not progress.
Skip the guilt ritual. Re-reading everything you didn’t do is a boss battle you can’t win on day one. Treat the restart like a new save file with carried-over XP: you keep the knowledge, you drop the debt. One 20-minute session plus one quick win is the whole protocol. Test it this week.
How do I make finishing the default instead of relying on motivation?
External structure beats willpower
Motivation is the novelty spike, and it fades on schedule. ADHD brains show altered dopamine reward signaling that biases attention toward immediate stimulation over delayed payoff (Volkow et al., JAMA, 2009), which is why “I’ll push through later” loses to any shiny new tab. Structure that exists outside your head wins: a checkpoint you wrote down at the start, a visible XP bar, a boss battle with a deadline attached.
Where gamified tools earn their keep, and where they don’t
Gamification works when it rewards the boring middle, not just the launch. XP for showing up on day nine, a streak that survives a setback, a quest that shrinks ambiguity into one next action — those are structural, not motivational. It fails when the game becomes the project: if you’re optimizing your task matrix instead of shipping, the tool is the new novelty spike.
Honest limits: SideQuest won’t restart a stalled project for you, and if you’re the person who deletes apps after 30 unused days, no XP curve fixes that. The default that matters is one checkpoint, written down, before the fun part starts.
How do the five checkpoints hold up in a real tool like SideQuest?
Where SideQuest maps cleanly to the checkpoints
The novelty spike is SideQuest’s home turf: XP, leveling, and Boss Battles are built for the dopamine hit that keeps you past day three. Framing a task as a quest reduces ambiguity, since a quest has a defined win condition. Daily consistency mechanics give you a quick win loop, so the near-done stall has a visible reward waiting on the other side.
Where it falls short
Mid-project boredom is the weak point. The gamification fires hardest at the start, and if the XP loop stops feeling fresh, you’re back to willpower for the long middle. Setbacks are also under-served: a Boss Battle that stops moving can sting enough to trigger an abandon, even though SideQuest never takes your progress away, and there’s no built-in restart ritual for a quest you dropped months ago. Run the 30-day test honestly: if SideQuest itself goes unused, delete it like any other tool.
FAQ
Is it normal to have a dozen unfinished projects, or is that a me problem?
Normal, and predictable. Projects die at the same five points every time: the novelty spike, the boring middle, ambiguity, the first setback, and the 90% stall. A graveyard of abandoned apps means you keep hitting the same walls, not that you lack discipline. Name where each one died and set a checkpoint there next time.
How long does the novelty phase usually last before I lose interest?
Roughly a week. The first few days run on the dopamine hit of a new tool or plan, then the spike drops somewhere around days 3 to 10 and the project becomes ordinary work. That is why most app-hopping happens in the first week: the tool did not fail, the novelty did. Plan your first checkpoint for day 7.
Should I force myself to finish something I’ve stopped caring about?
Not automatically. Run the deliberate-quit test in writing: does it still solve a problem you have, would you restart it fresh today, and is the remaining work specific enough to schedule? Two noes out of three means quit on purpose with a dated decision, otherwise you carry an open loop that drains energy from everything else.
What’s the difference between quitting and failing?
Failing is drift: novelty faded, no checkpoint forced a decision, and the project quietly stopped. Quitting is a deliberate act with a written if-then plan, like archiving a project after 14 skipped days. A deliberate quit is a completed task; drift is an open loop. Same outcome on paper, completely different effect on your next project.
Can a gamified app actually stop me from abandoning things, or is that just another novelty?
It can help, with limits. XP, streaks, and boss battles manufacture fresh dopamine triggers mid-project, which is what ADHD attention needs once novelty fades. But the gamification fires hardest at the start, the boring middle stays the weak point, and no app restarts a stalled project for you. The checkpoint you write down matters more.