Track How the Numbers Changed
Compare monthly datasets and surface what moved, what stayed flat, and what reversed
Comparing two versions of the same report by eye wastes hours and you still miss subtle shifts in the middle rows. Managers want the story of change, not two static snapshots.
That is the gap this drop closes. It is a data & analysis skill built for Kimi, and it takes about ten minutes to set up the first time. After that it runs in under a minute.
Who should use it
Analysts who receive updated datasets each period and need to brief managers on movement
How it works
The skill file does five things, in this order.
1. Upload both periods clearly labelled. Name files so the model knows which is baseline and which is current.
2. Ask for movement in each key metric. Tell it which columns matter and what counts as material change.
3. Request direction and magnitude. Get absolute change, percentage change, and rank shifts in one table.
4. Identify reversals and outliers. Flag anything that switched direction or moved beyond normal variance.
5. Generate a narrative summary. Ask for three sentences that a non-analyst can repeat in a meeting.
What comes back
A comparison brief that shows what grew, what shrank, what flipped, and where the data held steady, ready to discuss in five minutes
The mistake to avoid
Set thresholds for what counts as notable before you ask, or the model will highlight every tiny wobble. If your data has seasonal patterns, mention the expected range so real exceptions stand out.
Running it
Paste the prompt block into a new Kimi chat and attach your files in the same message. Kimi's long context means you should upload everything at once rather than in batches, splitting the input is what causes it to lose cross-references between documents.
Where this fits
On its own, one skill saves an hour a week. The compounding happens when three or four of them run in sequence on the same input, the same transcript that produces a scope of work also produces the follow-up email and the project brief. That is the point at which it stops being a prompt and starts being an internal tool. If you want that wired into the systems your team already uses, that is the work 67 Digital does.
In the file
I have two datasets for the same metrics across different time periods.
Baseline period: {month_year_baseline}
Current period: {month_year_current}
Key metrics to track: {list_of_metrics}
Material change threshold: {percentage_or_absolute_value}
For each metric, calculate:
- Absolute change
- Percentage change
- Direction (increase, decrease, flat)
- Rank change if applicable
Flag any metric that:
- Changed by more than the threshold
- Reversed direction compared to the prior trend
- Moved beyond {X} standard deviations from the mean
Present findings in a table, then write a three-sentence executive summary that states:
1. The biggest mover and its magnitude
2. Any reversal or unexpected stability
3. Overall pattern (broad growth, mixed signals, decline)
Highlight which changes are outside normal variance for {seasonal_context} and which align with expectations.