The Self-Running Weekly Review: Turn a Brain-Dump Into a Ranked Week
Empty your head into Claude and get back a week sorted by actual impact β not by whatever's shouting loudest in your inbox.
Every week starts the same way: a churning mental list of everything you're supposed to do, no clear sense of what actually matters most, and a creeping dread that you'll spend the week busy but somehow miss the thing that counted. So you default to whatever's most urgent or most annoying, and the important-but-quiet work keeps sliding.
The problem isn't that you don't work hard. It's that prioritising well is genuinely difficult β it means holding twenty competing things in your head and honestly weighing impact against effort, which is exactly the kind of thinking that falls apart when you're tired and the list is long. So it doesn't happen, and the loudest task wins by default.
This is the one where Claude earns its keep. You do the brain-dump β messy, unfiltered, everything. Claude does the sorting: clarifying, grouping, and ranking by real impact, then handing you back a week you can actually execute. You keep the judgement calls. It takes the overwhelm.
What You Need
- A free or Pro account at claude.ai β free tier works fine for this.
- 15 minutes, ideally at the same time each week (Sunday evening or Monday morning are the classics).
- Everything currently on your plate β work, life-admin, the lot. Half-formed is fine.
- Recommended: a Claude Project called "Weekly Review" so Claude remembers your goals and last week's list, and the review gets sharper each time.
Step 1: Set the frame β what actually matters right now
Before ranking anything, Claude needs to know what "important" means for you this season. Do this once (or update it monthly) at the top of your Project.
You are my sharp, no-nonsense chief of staff. Before we plan any week, I want you to hold onto my current priorities so you can rank my work against what actually matters β not just what's urgent. Here's what matters to me right now: - My top 2β3 goals this quarter (work and life): [FILL IN] - What a genuinely good week looks like for me: [FILL IN] - The things that feel urgent but usually aren't worth much: [FILL IN] - My hard constraints this week (fixed commitments, low-energy days, deadlines): [FILL IN] Store this as the lens you'll use to prioritise. If anything's unclear or contradictory, ask me up to 3 questions now. Push back on me if my stated priorities don't match where I'll clearly need to spend my time.
What just happened: you've given Claude the one thing generic to-do apps never have β a definition of what "important" means for you specifically, right now. Every ranking from here on is measured against your real goals, not against how loud a task happens to be.
Step 2: The brain-dump (do it badly, on purpose)
Now empty your head. Don't organise, don't filter, don't write full sentences. The whole point is to get it all out so Claude can do the sorting.
Here's my full brain-dump for the week β everything on my plate, in no order, work and personal mixed together. It's messy on purpose: [DUMP EVERYTHING HERE β tasks, worries, half-ideas, "I should reallyβ¦", errands, follow-ups. Don't tidy it.] Do this: 1. Turn my dump into a clean list of clear, concrete tasks. Split anything vague into real next actions, and merge duplicates. 2. Flag anything that isn't actually a task this week (someday-maybe, or a worry with no action) and set it aside separately so it's out of my head but off my plate. 3. Ask me about anything genuinely unclear rather than guessing. Don't rank it yet β just give me the clean list and the set-aside pile.
What just happened: the swirl in your head is now a clean list on a screen β and the stuff that was taking up mental space without actually being a task got named and parked. That alone drops the mental load. But the real value is next.
Step 3: Rank it by impact and build the week
Now the judgement. Claude weighs each task against your priorities from Step 1 and hands you an ordered plan.
Now rank my clean task list against the priorities you're holding for me. 1. Score each task on impact (how much it moves my real goals) and effort (time/energy required). Use a simple High/Medium/Low for each. 2. Surface the ONE task that would make the biggest difference this week β my "if I only do one thing" task. 3. Sort everything into: Do First (high impact), Schedule (important, not urgent β don't let these slide), Quick Wins (low effort, worth clearing), and Reconsider (low impact β should I even do these?). 4. Lay it across my week accounting for my constraints and energy, so I'm not stacking hard work on a day I said would be flat. 5. Be honest: tell me if I've planned more than is realistic, and what you'd cut. Give it to me as a clear, scannable plan I can start executing today.
What just happened: you've gone from a churning mental list to a ranked, week-shaped plan with a single clear priority at the top β the exact opposite of the "busy but adrift" default. And because Claude flagged what to cut, you're protected from the classic trap of planning a heroic week you'll never actually complete.
Bonus: The Friday close-out
A review that only looks forward misses half the value. Run this on Friday to learn from the week and feed the next one.
It's my Friday close-out. Here's what actually happened this week: [WHAT I FINISHED, WHAT SLID, WHAT SURPRISED ME]. Compare it to the plan we made. Tell me: what got done vs what I intended, what kept getting deferred and why that might be, one thing that went better than expected, and one specific adjustment to make next week's plan more realistic. Keep it honest and constructive, not a telling-off.
This is the loop that makes the whole system get smarter. Instead of repeating the same over-optimistic planning every week, Claude helps you notice your real patterns β what you always underestimate, what you always avoid β and adjusts. Two months of this and your weekly plans start actually matching reality.
Bonus: The mid-week rescue
Weeks fall apart by Wednesday. Use this to reset without spiralling.
My week has gone sideways. Here's where I'm at: [WHAT'S HAPPENED, WHAT'S NOW ON FIRE, WHAT I HAVEN'T TOUCHED]. Re-triage the rest of my week against my priorities. Tell me what genuinely still needs to happen, what can slip to next week without real cost, and the single most important thing to protect. Help me let go of the guilt about the plan changing β reworking the plan is the plan working, not failing.
Most people abandon their whole system the moment a week goes off-script. This turns a derailment into a two-minute reset β so one bad Tuesday doesn't cost you the entire week.
Claude didn't decide what matters to you β you did, in Step 1. All it does is take the exhausting cognitive work of sorting, weighing and ordering off your plate, so your energy goes into doing the work instead of agonising over what to do first. You set the priorities, you make the calls, you do the work. That's the whole philosophy: think better, not less. Offload the triage; keep the judgement. You were always the one steering the week β now you can see the road.
Ready to go deeper?
Build your own AI assistant in one afternoon.
A step-by-step guide to building a custom AI assistant that actually fits your life and work β no coding, no overwhelm, just a system you can use from day one.
Get the guide β $47Daily Stack
The tools behind the work.
The tools that have earned a permanent spot in my workflow.
Claude
Anthropic's flagship product and the tool that almost every resource on this site is designed for.
ManyChat
The easiest way to build automated Instagram DM flows, comment replies and lead capture without coding.
Planoly
My go-to for planning, scheduling and visually mapping out social content across platforms.