Starting is its own operation
There is a specific and very common ADHD experience: sitting in front of a task you have decided to do, that you are able to do, that you want done, and being unable to begin. Not distracted — you are looking directly at it. Simply not starting.
This is task initiation failing. It is a separate function from planning, from motivation, and from execution, and it can fail while all three of those work perfectly. That is why it feels so absurd from the inside, and why explanations aimed at motivation land so badly.
Once you accept that starting is its own operation with its own cost, the strategy becomes obvious: stop trying to supply more force, and start reducing the cost.
What makes a task expensive to start
Four properties reliably raise initiation cost. Most stuck tasks have at least two.
Vagueness. "Work on the report" has no defined first physical action. Your brain cannot begin an undefined thing, so it stalls looking for the entry point.
Size. The perceived weight of the whole task attaches to the moment of starting, even though starting only involves the first thirty seconds.
Emotional load. Tasks that carry shame, fear of judgement, or a history of failure attract avoidance that operates well below deliberate thought. This is why the task you have avoided longest is the hardest to start, independent of its actual difficulty.
Ambiguity about "done". If you cannot tell what finishing looks like, starting means entering something with no exit, which the brain treats as a threat.
Note that none of these is "not caring". Most stuck tasks are stuck precisely because you care.
Ten things that lower the cost
1. Shrink the first step until it is embarrassing.
Not "write the report" — open the document and type one bad sentence. Not "go to the gym" — put on the shoes. If the step still feels like a task, it is too big. The correct size is the one where refusing feels sillier than doing it.
2. Make the first step physical.
Cognitive first steps ("think about the structure") are almost impossible to initiate because nothing observable changes. Physical first steps ("open the file", "stand up") are much cheaper, and the body's movement pulls the rest along.
3. Define the entry point the night before.
Decide the exact first action while you have capacity, not at the moment you need it. Write it as a sentence: "Open budget.xlsx and fix the January column." The specificity is doing the work — you have removed the search for an entry point.
4. Start badly, deliberately.
Perfectionism is a major, frequently invisible driver of ADHD initiation failure. Explicitly plan to produce a bad first version. Write the worst possible opening sentence on purpose. The standard is what is blocking you, so remove the standard from the first five minutes.
5. Use another person's presence.
Body doubling — someone else present, doing their own work, not supervising you — reliably lowers initiation cost. Nobody fully agrees why; it works regardless. A silent video call with one person for two hours is the highest-yield intervention on this list.
6. Borrow momentum from a routine.
Tasks are much easier to start from inside a sequence that is already running than from a standing stop. This is the real argument for routines: not the individual steps, but that step four is nearly free because steps one to three built momentum.
7. Externalise the timer.
"I will work on this for fifteen minutes" without a visible timer becomes an unbounded commitment, and unbounded commitments are frightening to enter. A visible countdown makes the commitment finite and small.
8. Reduce the distance to zero before you need it.
Set up the environment in advance while capacity is high: file open, tabs closed, materials on the desk, gym bag by the door. At the moment of starting you should have to traverse no distance at all.
9. Say the first step out loud.
Verbalising the immediate next action — literally speaking it — helps some people bridge the gap between intending and moving. Cheap to test, works surprisingly often.
10. Attach it to something already happening.
Habit stacking: "after I make coffee, I open the document." An existing behaviour supplies the cue, so initiation happens in the slipstream of something already in motion.
What does not work
Willpower. Effort is not the limiting variable. Applying more of it produces exhaustion and no start.
Motivation-first thinking. Waiting to feel like starting means waiting for a state that arrives, at best, unpredictably. Action generates the state more often than the reverse.
Bigger consequences. Raising the stakes increases emotional load, which increases avoidance. It sometimes works at the very last minute via urgency — see below — but it degrades everything else in the meantime.
Guilt. Consumes the same executive capacity the task needs. Strictly negative.
About the deadline thing
Many people with ADHD find they can only start when a deadline is imminent. This is real: urgency supplies external pressure that substitutes for the missing internal initiation signal, and the work often gets done well.
It is also a poor system, because it requires a crisis to function, it removes any margin for the work being harder than expected, and it is exhausting.
The partial fix is manufacturing smaller, earlier, real deadlines: telling someone you will send them a draft on Thursday, booking a review meeting, working alongside someone with a shared end time. Fake deadlines you set for yourself mostly do not work, because you know they are fake. Deadlines involving another person mostly do.
When you are completely stuck
If nothing on the list moves you, stop trying to start the task and do this instead:
- Write down the exact next physical action, in one sentence. If you cannot, the task is too vague — that is the actual blocker, and defining it is now the task.
- Halve it. Then halve it again.
- Set a visible five-minute timer.
- Do only that, with explicit permission to stop when the timer ends.
Most of the time you will continue past five minutes, because the expensive part was the transition, not the work. And when you do stop at five minutes, you still started, which is the thing you were unable to do a moment ago.
Where an app helps
Not by reminding you. Reminders arrive from outside, demand an interruption, and get dismissed.
An app helps by making starting cheap: one tap from a widget to a running routine, the first step already defined and deliberately tiny, a visible external timer, and the sequence carrying you into steps two and three so momentum does the rest. That is the design Orbit is built around — routines start from the Lock Screen or a widget, show one small step at a time, and keep running on the Lock Screen so a lapse in attention costs a few seconds instead of the whole attempt.