Month-on-month change is (this month - last month) / last month. In a sheet with last month in B2 and this month in C2, type =(C2-B2)/B2 and format the cell as a percentage. To stop a zero in last month's cell from producing an error, use =IF(B2=0, "n/a", (C2-B2)/B2).
A household sheet with grocery and fuel
Put months down column A, grocery spending in B and fuel in D, with the change columns beside them. For the first month there is nothing to compare with, so leave it blank. From the second month:
- C3: =IF(B2=0, "n/a", (B3-B2)/B2), copied down. This is the grocery change.
- E3: =IF(D2=0, "n/a", (D3-D2)/D2), copied down. This is the fuel change.
- Format C and E as percentages with one decimal place.
A hypothetical four months:
| Month | Grocery | Change | Fuel | Change |
|---|---|---|---|---|
| January | ₹7,200 | - | ₹2,400 | - |
| February | ₹7,920 | +10.0% | ₹2,880 | +20.0% |
| March | ₹7,500 | -5.3% | ₹0 | -100.0% |
| April | ₹8,100 | +8.0% | ₹1,800 | n/a |
Grocery rose from ₹7,200 to ₹7,920, a rise of ₹720 on ₹7,200, which is 10.0%. The month with no fuel shows -100.0%, and the following month shows "n/a" because dividing by zero has no meaning; the guard keeps the sheet tidy instead of showing #DIV/0!.
IF or IFERROR: which guard to use?
IFERROR(formula, "n/a") is shorter, but it hides every kind of error, including ones caused by a typo or a text value in the cell. The IF version only covers the case you expect, which is a zero in the base month, and lets other problems show themselves. Use IF while building the sheet and IFERROR only when you are sure the data is clean.
How do I colour rises and falls automatically?
For spending, a rise is the thing you want to notice, so colour it warmly.
- Select the change cells, for example C3:C14 and E3:E14.
- In Excel, go to Home, Conditional Formatting, Highlight Cells Rules, Greater Than, type 0 and choose a light red fill.
- Repeat with Less Than 0 and a green fill.
- In Google Sheets, use Format, Conditional formatting, Greater than 0 for red and Less than 0 for green.
Text such as n/a is ignored by those rules, so it stays plain.
Where do the monthly totals come from?
If every expense sits in a list with a date in column A, a category in B and an amount in C, put the first day of each month in a header cell, say F1 holding 1-Jan-2026, and use:
- =SUMIFS($C$2:$C$500, $B$2:$B$500, "Groceries", $A$2:$A$500, ">="&F$1, $A$2:$A$500, "<"&EDATE(F$1, 1))
- Copy it across for each month, and change the category name or point to a cell holding it for fuel and others.
The argument order in SUMIFS is: the range to add, then each pair of a range to test and its condition. The EDATE function moves a date forward by whole months.
Why does a big swing not always mean something is wrong?
Month-on-month change is noisy. Festival months, an annual insurance premium, a month with a trip or a bulk purchase of rice can all produce a large percentage. Compare against the same month last year, with =(C14-C2)/C2 across a 12-month gap, and look at a three-month average to separate a one-off from a trend.
Checks before you read the column
- Each cell compares the right pair of months, not the same month twice.
- The percentage format is on, otherwise 0.1 looks like nothing.
- A zero base month shows n/a and not an error.
- Spending is entered as a positive number in both months, so a rise is a positive percentage.
Building the monthly totals by hand is the slow part. SpendixAI can import your bank statements and give you category totals by month through its reports, which you can compare in this same layout.
Useful ideas start with a clearer picture of your own spending.
Explore your reports


