Break it down: how to shrink a task until it starts

"Clean the garage." "Do my taxes." "Write the report." These don't fail to start because you're lazy. They fail to start because they aren't tasks — they're categories. A category has no first move, and a brain that can't find the first move won't begin. The fix isn't more willpower. It's a smaller task.

Breaking things down is the most-recommended ADHD advice on earth, and also the least useful — because nobody tells you how small is small enough, or what to do when the breaking-down is itself the part you can't start. Let's fix both.

Why a big task won't start

Working memory is the mental desk where you hold the steps of a plan while you carry it out. In ADHD it's commonly stretched thin — and an undefined task lands on that desk as one giant, undivided blob. Your brain tries to hold "do my taxes" and quietly runs the math on everything it might involve: the forms, the logins, the missing receipt, the number that might be bad news. That's not one step. That's twenty, all firing at once, with no order.

Cognitive-load research has a plain name for this: when the load exceeds what working memory can hold, performance doesn't just dip — it stalls. The task feels heavy because it literally is, in the one place you can't add capacity on demand. So the move isn't to push harder against the blob. It's to make the thing on the desk smaller.

The size of the step is the whole game

Here's the rule most advice skips: a step is small enough when you can picture yourself physically doing it in the next two minutes. Not "start the taxes." Not even "gather the documents." Try: "open the tax folder on the desktop." That's it. That's the whole step.

If you read a step and still feel a flicker of "…but how," it's not a step yet — it's a smaller category. Cut it again. You're looking for the very next physical action: something with a verb, an object, and no hidden decisions buried inside it.

Why smaller steps actually move you

This isn't a productivity trick — it's how motivation is built. In a classic study, children stuck on math did far better when a distant goal was split into proximal subgoals — small, near targets — than when they aimed at the far one. The subgoals didn't just organize the work; they raised self-efficacy: each small win was proof the next was possible (Bandura & Schunk, 1981).

That's the quiet mechanism under breaking things down. A blob offers no evidence you can do it — only the size of what you can't. A first step you actually finish delivers a real signal: begun, and survivable. Momentum in ADHD is rarely about discipline. It's about getting one honest piece of proof onto the board.

Name the trigger, not just the step

One upgrade makes a broken-down step dramatically more likely to happen: attach it to a when and a where. "If it's still light when I get home, I'll open the tax folder before I sit down." Psychologists call these implementation intentions — simple "if-then" plans that pre-decide the moment of starting. Across hundreds of studies they reliably raise follow-through, because they hand the hardest part — deciding to begin, in the moment — to a plan you already made (Gollwitzer, 1999).

The point isn't to schedule your life to the minute. It's that "later" is where broken-down steps go to die. A trigger gives the step a door.

When your brain won't do the breaking down

Here's the trap nobody warns you about: breaking a task down is itself a task. It needs the exact planning function that's offline in the moment you're frozen. Telling a stuck brain to "just break it into steps" can be like telling someone who can't stand up to "just take the stairs."

So don't make decomposition another thing you have to be capable of first. Externalize it:

• Say it out loud. Describe the task to a person, a voice note, or an empty room. The mouth often finds the first step the frozen planner can't.
• Borrow a brain. Ask someone — or a tool — "what's the very first thing?" You don't need their whole plan. You need step one, so you're not the one generating it from zero.
• Answer, don't invent. Reacting to "is it this, or this?" is far easier than producing an answer from a blank page. Turn planning into picking.

None of this is cheating. Offloading the planning step to something outside your overloaded working memory is exactly the right move — it's the accommodation, not the failure.

A task you can't start isn't too hard. It's too big. Shrink it until the next action is something your hands could do in two minutes — and if you can't do the shrinking, let something else hold the pieces for you. — The whole method, in two sentences

What it looks like in practice

A few blobs, cut down to a real first step:

1. "Answer that email." → Open the email and read it once. That's all. The reply is a separate step for later.
2. "Clean the kitchen." → Pick up the three things nearest the sink. Not the kitchen. Three things.
3. "Write the report." → Open a blank doc and type the one ugly sentence you'd say out loud if someone asked what it's about.

Notice none of these finishes the task. That's the point. You're not trying to do the thing — you're trying to begin it, because beginning is the executive function that was actually broken. Once you're moving, the next step is almost always easier to see than it was from a standstill.

How Beginary uses this

This is exactly what AI Breakdown does in the app. You hand Beginary the thing you've been avoiding — typed or spoken — and it splits it into small, doable steps and hands you the first one, marked start here. You don't have to generate the plan from a blank page while frozen; you react to one that's already there. And because the first step is sized to actually be startable, the app can drop you straight into it — no run-up required.

Breaking it down is the move. Beginary just makes sure you're never the one who has to do the breaking down alone.