Smooth noisy figures with a rolling average

Prompt that chooses a window, calculates a rolling average and explains the trend that remains. Paste it into an AI tool of your choice, which then smooths the figures.

Topic: Spreadsheets & data
Type: Prompt
Level: Intermediate
Time: about 15 min

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

Result preview

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

| Date | Orders | 7-day average | | Mon 3 | 120 | 104 | | Tue 4 | 96 | 106 | | Wed 5 | 88 | 107 | Window: 7 days (strong weekday pattern), trailing Report: orders are rising slowly, about 3% in two weeks; the average reacts to changes with a delay of a few days.

Fill in

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

FiguresWhich figures?

RhythmWhich pattern repeats?

QuestionWhat do you want to know?

Your prompt

Help me read a noisy series of figures with a rolling average.
Figures: [Figures]
Rhythm: [Rhythm]
Question: [Question]
Approach: 1. Describe the raw series briefly: level, spread and visible patterns by weekday or month. 2. Propose a window length that matches the rhythm, for example seven days for weekday effects, and explain the choice. 3. Calculate the rolling average and say whether it is centred or trailing. 4. Describe the trend that remains and where it changes direction. 5. Mark points where the average hides a real event, such as an outage or a campaign. 6. Write two sentences for a report that state the trend and its limits.
Output: table with date, raw value and rolling average, the chosen window with reason, and the two report sentences.

Take particular care to choose the window from the rhythm of the data and state what the average hides.

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.

Help me read a noisy series of figures with a rolling average.
Figures: [Figures]
Rhythm: [Rhythm]
Question: [Question]
Approach: 1. Describe the raw series briefly: level, spread and visible patterns by weekday or month. 2. Propose a window length that matches the rhythm, for example seven days for weekday effects, and explain the choice. 3. Calculate the rolling average and say whether it is centred or trailing. 4. Describe the trend that remains and where it changes direction. 5. Mark points where the average hides a real event, such as an outage or a campaign. 6. Write two sentences for a report that state the trend and its limits.
Output: table with date, raw value and rolling average, the chosen window with reason, and the two report sentences.

Take particular care to choose the window from the rhythm of the data and state what the average hides.

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.
Paste the series as two columns, date and value; at least three times the window length gives a stable picture.

Why does this work?

Single days jump around for reasons nobody controls. A window that matches the rhythm of the data shows the direction without inventing a trend.

Tips

A rolling average lags behind; for decisions about a single day, look at the raw value too.

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

Smooth noisy figures with a rolling average

Result preview

Fill in

Your prompt

Where this prompt works

How to use it

Why does this work?

Tips

Also a good fit

Find cardinality explosions in metrics New

Estimate customer lifetime value simply New

Read a retention cohort table New