Jan 9, 2026
8 min read
The 45-Minute Founder Weekly Review That Actually Sticks
My entire weekly review fits in 45 minutes. Four steps to clear inboxes, check metrics, set priorities, and kill everything else.
I’ve tried every weekly review system out there. David Allen’s full GTD sweep. The Bullet Journal migration ritual. Elaborate Notion dashboards with rolling twelve-week views. They all worked for about three weeks, then I’d skip one Sunday and never come back.
The problem wasn’t discipline. The problem was time. A review that takes 90 minutes competes with everything else in your life. A review that takes 45 minutes fits into a gap. That’s the difference between a system that sticks and one that dies.
I’ve been running the same 45-minute weekly review for over two years now. It has four steps, a hard time cap on each one, and it produces exactly one artifact: next week’s top three priorities. Everything else is a means to get there.
Why most weekly reviews fail
They fail because they’re designed for a version of you that has unlimited patience on a Sunday evening. The classic weekly review asks you to process every open loop, review every project, update every list, and then do a “higher altitude” reflection on your goals and values. That’s beautiful in theory. In practice, it’s a 2-hour commitment that feels like homework.
Founders especially get burned by elaborate reviews because our work doesn’t break into neat projects. Half of what I dealt with last week doesn’t map to any task list — it was a text from a vendor, an idea I had in the shower, a problem that surfaced in a customer call. Trying to capture and process all of that in a structured review is a losing game.
The fix is constraints. Forty-five minutes total. Four steps. A timer for each one. If you don’t finish a step in the allotted time, you move on anyway. The review is about direction, not completion.
The 4-step flow
Step 1: Clear inboxes (15 minutes)
Set a timer. Open every inbox you have — email, Slack DMs, voice memos, your notes app, texts you flagged, screenshots you saved. The goal isn’t to respond to everything. The goal is to get every input to zero or to a task list.
For each item, you have three options:
- Do it if it takes under two minutes
- Add it to your task list if it needs more time
- Delete or archive it if it doesn’t matter
I process roughly 40-60 items during this step. Most get archived. Maybe 8-10 become tasks. The rest were noise I was carrying around as mental weight without realizing it.
The key discipline: do not start working on anything that takes more than two minutes. You’re processing, not producing. The moment you open a Google Doc to “quickly draft” a response, the review is dead.
Step 2: Review metrics (10 minutes)
You don’t need a dashboard with 47 KPIs. You need 3-5 numbers that tell you whether the business moved forward last week. Mine are:
- Revenue (trailing 7 days vs. previous 7)
- Email list growth (net new subscribers)
- Site traffic (sessions, top pages)
- Cash position (Mercury balance, quick glance)
- Pipeline count (active deals or projects in motion)
I pull these from the same places every week — Shopify, ConvertKit, Fathom, Mercury. The entire step takes about 10 minutes because I’m not analyzing anything. I’m scanning. I’m looking for one thing: is anything off?
If revenue dropped 30%, that’s a signal. If the email list flatlined, that’s a signal. If everything looks normal, I note “metrics steady” and move on. The weekly review is not the place to diagnose problems. It’s the place to notice them so you can schedule time to dig in during the week.
Step 3: Pick your top 3 priorities (10 minutes)
Look at your task list — the one you just added to in Step 1 — plus anything carried over from last week. Now ask: If I could only finish three things next week, which three would move the business forward the most?
Not the most urgent. Not the ones with the closest deadline. The ones with the most leverage.
This is the “if nothing else gets done” test. If next week goes sideways — someone quits, a product breaks, you get sick on Wednesday — and you still managed to finish these three things, would the week count as a win?
Write them down. Be specific. Not “work on marketing” but “publish the Q1 case study and send it to the email list.” Not “deal with hiring” but “make a final call on the design contractor and send the offer.” Vague priorities don’t get done.
I’ve experimented with top 5, top 1, and no limit. Three is the number. One feels artificially constraining and makes you anxious about what you’re ignoring. Five dilutes focus. Three forces a real decision about what matters without pretending you’re a machine.
Step 4: Kill or defer everything else (10 minutes)
This is the step most people skip, and it’s the most important one. Look at everything on your list that didn’t make the top three. For each item, decide:
- Defer it — move it to a “next week” or “someday” list with a specific date to reconsider
- Delegate it — assign it to someone else with a clear brief
- Kill it — accept that it doesn’t matter enough and delete it
The point of this step is to go into Monday with a clean list. Not a list of 47 things with three starred. A list of three things, period. Everything else is handled, deferred, or gone.
I find that about 20% of my “important” tasks from the previous week get killed here. They felt urgent when I wrote them down but a week later, nobody noticed they didn’t happen. That tells you everything about their actual importance.
When to do it
I’ve tried Sunday evening, Sunday morning, Friday afternoon, and Monday at 6 AM. My answer: Sunday around 5 PM.
Friday afternoon doesn’t work because I’m still in execution mode and can’t evaluate the week with any distance. Monday morning is too late — you’re already reacting. Sunday morning is fine but I found I’d procrastinate all day knowing it was hanging over me.
Sunday at 5 PM gives me enough distance from the work week to see clearly, and it’s late enough that I’m naturally shifting into “prepare for tomorrow” mode anyway. I finish by 5:45, cook dinner, and go into Monday knowing exactly what matters.
If you hate the idea of working on Sunday, do it Friday at 4 PM. The specific time matters less than the consistency. Pick a slot and defend it like a meeting with your most important client — because it is.
What to do with the output
Your top three priorities need to live somewhere you’ll see them Monday morning. I keep mine in a pinned Slack message to myself and in the Sunday System note I’ve written about before. Some people use a sticky note on their monitor. The format doesn’t matter. The visibility does.
I also keep a running log of every weekly review in a single Notion page — just the date and the three priorities. After a few months, you can scroll back and see patterns. You’ll notice which priorities keep showing up week after week without getting done (a sign you should either commit fully or kill them) and which ones you knocked out fast (a sign of what you’re actually good at and energized by).
How AI handles part of this now
One evolution worth noting: I’ve started using an AI agent to handle the first pass of Step 1 and Step 2. It pulls my unread emails, Slack threads, and key metrics into a single summary every Sunday morning. By the time I sit down at 5 PM, the raw material is already organized.
This cut my review from 45 minutes to about 30. I still do Steps 3 and 4 myself — priority-setting and ruthless pruning are judgment calls I don’t want to outsource. But the processing and data-gathering? That’s exactly the kind of repetitive work AI agents are built for.
The real point
A weekly review is not a productivity ritual. It’s a forcing function for honesty. It makes you look at what you said mattered last week and compare it to what you actually did. That gap — between intention and action — is where most founders lose months without realizing it.
Forty-five minutes. Four steps. Three priorities. Do it this Sunday. If you already have a morning routine that works, this is the weekly companion to it. And if you want to go deeper on annual cycles, I wrote about my year-in-review system that builds on the same principles.
The act of closing loops is what creates momentum. Not planning. Not reflecting. Closing.
Written by
Cathryn Lavery
Cathryn built and sold BestSelf, bought it back from private equity, and still runs it. She writes Little Might so she doesn't have to keep these lessons in her head.
Related reading
-
Apr 2, 2026
The Ops Layer That Keeps Your OpenClaw Agents Alive
-
Mar 25, 2026
Claude Code Tutorial: Hooks, Commands, Plugins, Skills, and MCP Explained
-
Mar 15, 2026
How I built my wife a personal AI assistant on OpenClaw (and what actually took time)
-
Mar 9, 2026
Everything You Need Before Your AI Employee Starts
-
Mar 4, 2026
Turn a Mac Mini into Your First AI Employee in 20 Minutes