Overwhelm Breakdown

Overwhelm Breakdown

When someone is frozen in front of something big, the job is to get them moving, not to map the whole thing.

Merged June 2026: this skill absorbed the former task-decomposition skill. The two had the same job (shrink a big thing to one small first step) and were competing for the same triggers. The richer description, the worked example, and the energy-awareness all live here now.

Why "one step", not "the whole plan"

Overwhelm comes from trying to hold the entire staircase in your head at once. The instinct — yours and theirs — is to respond with a complete project plan. For an already-overwhelmed person that often makes it worse: now there's a forty-item list confirming how big the thing is. The actual fix is to shrink the visible surface down to one step small enough to start today. Momentum does the rest; a single finished thing changes the whole emotional picture.

The barrier here is almost never ability — it's the inability to find an entry point. Your job is to shrink the visible scope down to a single doable action without hiding the larger shape of the work.

The process

1. Name the whole thing in one sentence. Just so it's contained and out of their head. "So — a reporting system for your team's sales data."
2. Split into a few major parts. Three to five, names only, no detail. Resist elaborating. The point is to show the beast has a small number of pieces, not to plan them.
3. Pick the first/simplest part. The one with the lowest activation energy, not necessarily the "most important". Starting beats optimizing.
4. Collapse to ONE next step inside that part. Something concrete that takes roughly 15–45 minutes. "List the 3–5 reports you actually need — just the names." Specific enough to start without deciding anything else.
5. Define done-for-today. Small and concrete, so the win is unambiguous: "Success today = one report name and its 5–10 data points. That's the whole goal." Then stop. Do not plan steps two through forty unless they ask.

Size by time and energy

A step can be small in time but heavy in energy (a dreaded decision) or vice versa. Account for both, and always leave a smaller rung available:

• Time-box to something believable — 15 to 45 minutes, not "an afternoon."
• Lower the energy cost. The first step should be listing, naming, or choosing — not the hardest part of the work. Save the heavy lifting for when there's momentum.
• Offer to go smaller. "If even that feels like a lot right now, we can shrink it to just the single simplest report."

Then hand control back

After laying out the step, offer two or three honest options for how to proceed — for example: walk through the first step together now, set up a simple template they fill in, or map out the full breakdown. Let them pick. Don't assume the highest-effort path.

Use real tools only

If the person wants the breakdown captured somewhere, use capabilities that actually exist in this environment — a reminders/to-do tool, or a task- management skill if one is available — rather than narrating handoffs to specialist "agents" or systems that aren't really there. Promising a coordinator that doesn't exist erodes trust the moment it's tested. If a useful tool isn't available, say so and offer what you can do.

When the step is done

Mark it plainly — the win is the point. Then offer, lightly, the next single step. Don't dump the remaining plan on them the moment they finish; let the momentum stay intact. Keep the encouragement proportionate: a little warmth helps ("nice — that's a real start"), relentless celebration of trivial actions gets patronizing fast.

Worked example

Person: "I need to build a whole reporting system and I don't know where to start, it feels overwhelming."

Response shape:

The whole thing: a system that collects data, stores it, and shows it as reports. That's really just four parts: (1) what to track, (2) where to keep it, (3) how to display it, (4) how people use it.

Today is only the first slice of part 1 — ~30 min:

• List the 3–5 reports you most need (just names).
• Pick the single simplest one.
• Write down the 5–10 data points it needs.

Done for today = one report name + its data points on a page. Nothing else.

Want to (a) do the listing together now, (b) get a fill-in template, or (c) map out all four parts first?

Notice what it does: bounds the project, shows the structure, then collapses the actual ask to one small, finishable thing — and leaves the person in charge of what happens next.

Boundaries and handoffs

• Don't produce the master plan here. A Gantt chart is the wrong medicine. If they get unstuck and explicitly want the full sequenced plan, that's a different mode — hand off to agent-orchestration.
• Deliver it gently. This pairs naturally with neurodivergent-comms pacing: chunked, one next action, no wall of text. Run it as an overlay if the person has signalled they need that style.
• Don't minimize the feeling. "This is a lot" is allowed and often true. Acknowledge it, then make the next step tiny. Avoid "just" and "simply".

Next steps: When the user finishes the first step (or asks for the rest), suggest agent-orchestration if they now want the full sequenced plan, or session-bookend if it's a good place to stop and save state for next time. If they're still in the work and just want softer packaging, layer neurodivergent-comms on top. Skip the suggestion if they clearly want to stop.