ANALYTICS & REPORTING

Monthly report narrative

Use when the numbers for the month are gathered and you need to write the commentary around them.

monthly-report-narrative.md

Download .md

You are writing the narrative section of a monthly performance report for {{CLIENT_OR_TEAM}}.

Numbers for {{MONTH}}:
{{METRICS_WITH_COMPARISONS}}

Known events during the period (launches, outages, campaigns, seasonality):
{{KNOWN_EVENTS}}

Write the narrative in this structure:
- What happened (4 to 6 sentences, the actual movement in the numbers)
- Why we think it happened (each claim tagged Confident, Likely, or Unclear, with the evidence in brackets)
- What we did about it last month
- What we will do next month (3 items, each with an owner placeholder and a date)

Constraints:
- Every percentage must be traceable to a number I gave you. If you want to state something you cannot support, write "we cannot verify this from the data we have".
- Report declines in the same tone as growth. Do not soften a drop with a cherry-picked secondary metric.
- No em dashes. No "in today's fast-paced world" style openers. No "it is not just X, it is Y" constructions. No closing paragraph that restates the opening.
- Plain English. A finance director who does not use our tools should understand every sentence.
- Under 500 words.

Fill in before running

Replace each placeholder with your own detail. The more specific you are, the less the model invents.

{{CLIENT_OR_TEAM}}
{{MONTH}}
{{METRICS_WITH_COMPARISONS}}
{{KNOWN_EVENTS}}

Getting a better result

Give it last month's narrative too, so it can say what changed rather than starting fresh.
Include the events list even when it feels obvious - most bad reporting is a real cause attributed to the wrong thing.
Ask it to list what it would need to move an Unclear tag to Confident.

Questions about this prompt

When do I use this rather than writing the commentary myself?

When the numbers are settled and the remaining work is saying what they mean. It structures a reading you have already done into what happened, why, what we did and what is next. If you cannot fill in the known events field, you are not ready for it and should do the analysis first.

What do I need in front of me?

The month figures with their comparisons, and an events list covering launches, outages, campaigns and seasonality. That events field is the one people skip, and it is why so much reporting attributes a real movement to the wrong cause. Last month's narrative helps too, so it can say what changed rather than starting fresh.

What comes back?

Four sections, with every causal claim tagged Confident, Likely or Unclear and its evidence in brackets beside it. Those tags are the useful part, because they stop a guess reading like a finding. The whole thing runs under 500 words in language a finance director who never opens your tools can follow.

What is the mistake that costs me here?

Letting an Unclear tag stay Unclear month after month. Ask what evidence would move it to Confident and turn that into a tracking ticket. The other one is softening a decline with a better looking secondary metric, which the prompt refuses to do and which your reader notices the following month anyway.