Compare monthly figures fairly with last year

Prompt that compares monthly figures with last year and removes working days, holidays and one-off effects before drawing a conclusion. Paste it into an AI tool of your choice, which then runs the year-over-year comparison.

Topic: Spreadsheets & data
Type: Prompt
Level: Intermediate
Time: about 8 min

Results depend on the tool, the model and what you enter.

Result preview

Example: written by our editors, not real AI output.

Orders compared with last year Month · this year · last year · change · per working day March · 1,840 · 2,010 · −8.5% · +0.3% (21 instead of 23 working days, Easter fell in March this time) April · 2,150 · 1,930 · +11.4% · +5.8% (20 instead of 19 working days) May · 1,990 · 2,240 · −11.2% · same working days One-off effect: summer promotion in May last year with about 300 orders. Without it: 1,990 vs 1,940 = +2.6%. Conclusion: the drop in March is a calendar effect. April is ahead per working day too. May is only missing the promotion. Too small to interpret: March per working day (+0.3%). Open question: did last year's promotion pull orders forward from June?

Fill in

What you type stays in this browser window. We store and send nothing.

MetricWhich figure are you comparing?

ValuesPaste the monthly values for both years, for example as a table

SpecialWhat was different in one of the two years?

Your prompt

Compare the following monthly values with the previous year.
Metric: [Metric]
Values per month for this year and last year:
<values>
[Values]
</values>
Known special circumstances: [Special]
Approach: 1. Calculate the change per month in absolute terms and in percent. 2. Check calendar effects: number of working days, moving holidays such as Easter or Whitsun, bridge days. Where it fits, also calculate per working day. 3. Separate one-off effects such as promotions, outages or price changes and state how much they distort the picture. 4. Name which changes remain after the adjustment and which are too small to interpret. 5. Finish with a three-sentence conclusion and one question the data cannot answer. Only use the values given and show the calculation briefly.

Take particular care to remove calendar and one-off effects before drawing conclusions and not read meaning into small differences.

Before you answer, check: Are all instructions met? If key information is missing or a placeholder in square brackets is still in the text, ask instead of guessing. Mark assumptions as assumptions.

Compare the following monthly values with the previous year.
Metric: [Metric]
Values per month for this year and last year:
<values>
[Values]
</values>
Known special circumstances: [Special]
Approach: 1. Calculate the change per month in absolute terms and in percent. 2. Check calendar effects: number of working days, moving holidays such as Easter or Whitsun, bridge days. Where it fits, also calculate per working day. 3. Separate one-off effects such as promotions, outages or price changes and state how much they distort the picture. 4. Name which changes remain after the adjustment and which are too small to interpret. 5. Finish with a three-sentence conclusion and one question the data cannot answer. Only use the values given and show the calculation briefly.

Take particular care to remove calendar and one-off effects before drawing conclusions and not read meaning into small differences.

Before you answer, check: Are all instructions met? If key information is missing or a placeholder in square brackets is still in the text, ask instead of guessing. Mark assumptions as assumptions.

This is an instruction for an AI, not the result itself.

Where this prompt works

These AI tools are a good fit for this prompt:

How to use it

Fill in the fields or tap a suggestion. Leave out names, addresses and other personal data.
Copy the finished prompt into an AI tool of your choice.
Check the working days for your region, because public holidays differ by country and region.

Why does this work?

A month with two fewer working days quickly looks like a slump. Removing calendar and one-off effects first means you only interpret real changes.

Tips

With a strong weekly rhythm, as in retail, comparing the same weekdays is often more meaningful than the calendar month.

You may use our prompts freely, including at work. Please do not republish them as your own collection. We give no warranty for AI results; check them yourself. Terms of use

Compare monthly figures fairly with last year

Result preview

Fill in

Your prompt

Where this prompt works

How to use it

Why does this work?

Tips

Part of these agents

Also a good fit

Simple forecast with a range from monthly figures New

Plan a pivot table for a question New

Check the unit of analysis in tests New