Q&A

What is a good AI prompt for writing a weekly work report?

November 1, 2026 · Brian Arfi Faridhi

Short answer

A good weekly report prompt has three parts: context (who you are and who reads the report), format (a fixed structure for the output), and raw input (your messy notes pasted in). Add one rule: the AI may not invent numbers. Then check every figure yourself before sending it to your boss.

I get this question a lot. The answer is short. A weekly report is a perfect job for AI because the pattern is the same every week. Only the content changes.

The most common mistake is typing "write my weekly report" and nothing else. The AI does not know where you work, who reads the report, or what happened. You get a generic report that could belong to anyone, so it is useful to no one.

Here is the template I use. Replace the brackets:

CONTEXT
I am a [role] at [team/company]. This report is for [manager/stakeholder]
who reads fast, two minutes at most. Tone: direct and honest.

FORMAT
Use this order:
1. Summary of the week, three sentences max
2. What got done, short bullets
3. What is still in progress, with expected finish dates
4. Blockers and the help I need
5. Priorities for next week, three max

RULES
Use only the data below. Do not add numbers or facts.
If data is missing, write "needs checking" in that section.

RAW DATA
[paste daily notes, task list, metrics, meeting notes]

What each part does

Context sets the tone. A report for a boss who reads between meetings is different from one for a team that needs detail. Name the reader and the AI adjusts length and style.

Format is your call. Do not let the AI choose. The same structure every week helps your manager scan faster, and you never rethink the layout.

Raw data decides the quality. Paste it as is, mess included. AI is good at cleaning up piles of notes. If you give it nothing, it will fill the gaps on its own.

Check the numbers

Do not skip this. AI can flip two digits or calculate a percentage from the wrong base, and it will sound completely confident. The rule "use only the data below" lowers that risk. It does not remove it.

Before you send, compare every number in the report with its source. It takes a few minutes. You are the one sending it, so you are the one accountable.

Make it stick

Save the prompt as a template. Next week, swap in new raw data. Want comparisons? Paste last week's report and ask the AI to highlight what changed.

This is the same pattern behind the 8-channel content distribution system I run on my own: raw input goes in, the format is fixed, AI drafts, I check. The principle is identical whether it is content or a work report.

Helping companies save money is an old habit of mine, from long before AI was trendy. As a product leader I drove more than 4 million US dollars in savings per year through a range of initiatives: process improvements, cost optimization, and more efficient product decisions. Automation was only one of them. Back then, that kind of leverage required a senior seat, an engineering team, and expensive systems. Today AI is the sharpest tool for that same habit, and it can be trained into everyone on the team. A weekly report is just the easiest way in.

If you want your whole team working like this, see my corporate page for how I structure the training.