How to Write a Weekly Status Update Your Manager Acts On

Most weekly updates list what you did. Put the decision your manager needs in line one: the prompt, a before and after, and where it goes wrong.

On this page

More responsibility every year. Same title. And every Friday, an update that lists everything you did. The textbook advice on status updates is about completeness: what got done, what is in progress, what is blocked. Follow it and you get an accurate list that reads as effort. The version that changes how people above you read your work is shorter and starts somewhere else: the first line is the decision your manager needs to make this week, followed by your recommendation. Same work. But now you sound like the person who decides.

The step: the decision in line one

Before you send this week's update, paste it into your AI assistant with one line:

TEXT
Rewrite this so the first line is the decision my manager needs to make.

Then read the first line it gives you and ask yourself one question: would I defend this in a meeting tomorrow? If yes, send it. If the line is wrong, fix the line, not the prompt. That correction is the part of the job only you can do.

If AI finds no decision in your update, that is an answer too. Some weeks are just progress. Then say so first: "No decision needed this week. On track for the March start." Your manager reads one line and moves on, which is what they wanted from the update anyway.

Before and after

An illustrative example, not a real update:

Before

  • Worked on the vendor review
  • Met with procurement and IT security
  • Collected pricing from three vendors
  • Next week: continue the review

After

We need to choose a vendor by Friday to keep the March start. I recommend B: it is the only one that passed the security review, and it sits in the middle on price. If you agree, I'll tell procurement on Monday.

Details: met procurement and IT security this week; pricing from all three vendors is in the attached sheet.

Nothing in the second version is new work. The vendors, the meetings and the pricing were all in the first one. What changed is that the second version did the synthesis instead of leaving it to the reader.

Why it works

A list of activity makes your manager do the thinking. They have to work out what the meetings and the pricing mean, and whether anything is needed from them. Most won't, not because they don't care but because your update is one of many. The synthesis you skipped is exactly the senior part of the job, and the update is the one place where people above you see whether you can do it.

A decision gets a reply. A list gets "thanks". "I recommend B, do you agree?" can be answered in one word. That reply is the start of being included earlier, which is how the next decision reaches you before it is made.

Over a few months, your updates become a record. Ten weeks of recommendations is evidence of judgment that no list of tasks can be. It is the same habit as a decision log, applied to your own work. When someone asks why you should lead the next thing, the record answers before you do. The next step, ending every update with a recommendation, makes that record explicit.

Where it goes wrong

  • Inventing a decision. If there isn't one, don't manufacture one to fill line one. "No decision needed" is honest and earns trust for the weeks when there is one.
  • Letting AI make the recommendation. AI can reorder what you wrote. It cannot know how your work really happens: that your security team will never sign off on vendor C, or that your manager has already promised a date. Check the first line against what you know.
  • Deciding above your level. Write a recommendation, not a decision taken, unless the decision is yours to take. "I recommend B" invites a yes; "We are going with B" invites a correction.
  • Burying the ask in politeness. "Just wanted to flag that we might potentially need to look at…" is still the old update. One sentence, the decision first.
  • Pasting what shouldn't leave the building. Use the assistant your company approves, or take out client names, amounts and anything confidential before you paste.

What I do

I built Delivery Sheet around the same idea, for requests instead of updates: every incoming ask has to end in one of four decisions instead of a reflexive yes. A request isn't ready until it names what has to be decided, and the same is true of an update. If you want the longer version for project requests, this is how I turn a Slack request into a decision-ready initiative, and the free Project Intake One-Pager turns a raw request into a decision-ready page.

Try it on this week's update before you hit send.

Share

Frequently asked questions

(3)

What should a weekly status update to your manager include?

Start with the one decision your manager needs to make, or the one you made and want them to know about, with your recommendation and a line of why. Then the risks that could change that decision. Then the progress, shorter than you think. If nothing needs deciding this week, say that in the first line; it is useful information too.

How long should a weekly status update be?

Short enough that the first three lines carry everything that matters. The rest is reference for anyone who wants detail. A long update is not wrong, but the length belongs below the decision, not in front of it.

Can I use AI to write my status update at work?

Yes, for the reordering and the tightening, with two conditions. Use the assistant your company approves, or take names, amounts and client details out first. And check the recommendation yourself: AI can move the decision to the top, but it cannot know which option your team can actually live with.

The daily step goes out on LinkedIn and X.