Converting an American Express India statement from PDF to Excel is mostly about preparation and verification. Look for a CSV or Excel download first, convert the PDF only if you must, then turn the result into one tidy table with real dates, numeric amounts and purchases and payments clearly separated. Check it against the statement's total due before you rely on it.
What should I try before converting the PDF?
Check whether the statement area offers other download formats. If it does, take the CSV or Excel version, because it needs far less repair. If you can only get the PDF, keep it as your master copy and work on a copy of the data.
We do not know the order of columns on American Express India statements. Open yours and note what each column holds before editing anything.
Copy, paste or convert?
For a one-page statement, copying lines straight into Excel is quick, but you may find a whole line in one cell. Use Data, Get Data, From File, From PDF in recent Excel versions, or a converter you trust. Statements contain your spending history, so avoid free sites that you know nothing about. Google Sheets has no built-in PDF import, so a conversion usually happens in another tool first.
How do I handle a protected file?
If the PDF asks for a password, follow the instruction that came with the statement. Open it on your own device. A converter that cannot open locked files will return nothing or garbage, which is a sign to save an unlocked working copy for yourself and delete it afterwards.
How do I clean the table?
- Remove everything above the column headings: account summary, offers and address.
- Delete page headers that repeat, and the totals or summary block at the end.
- Convert dates to real dates and sort by date.
- Check that every amount is a number. =COUNT(D2:D300) counts the numeric cells, and =COUNTA(D2:D300) counts the non-empty ones. They should be equal.
- Create one signed amount column, with purchases positive and payments and refunds negative, so a sum of the column gives the change in what you owe.
How do I separate payments and refunds from purchases?
Add a Type column. A simple test is the sign: negative or marked credit means payment or refund. To tell a payment from a refund, read the narration. Payments usually mention a payment received. Refunds mention a merchant. In a spare column, a formula such as =IF(D2<0,"Credit","Purchase") sorts rows into two groups that you can then filter.
| Row | Sign | Treat as |
|---|---|---|
| A grocery order | Positive | Spending |
| A bank payment toward the card | Negative | Not spending |
| A returned item from the same merchant | Negative | Reduces that category |
| A fee from the issuer | Positive | Spending, bank charges |
How do I verify it?
Say last cycle's balance was ₹5,500, you paid ₹5,500, new purchases came to ₹31,800 and refunds were ₹2,300. The new balance is 5,500 - 5,500 + 31,800 - 2,300 = ₹29,500. Add the signed column in your sheet and you should get 31,800 - 2,300 = ₹29,500 of net new charges. Match that to the statement before continuing.
How do I tag each row with a month?
Because billing cycles rarely match calendar months, add a helper column that assigns each row to the month of its transaction date. With dates in column A, put this in a new column: =TEXT(A2,"yyyy-mm"). Sorting or filtering on that column then groups spending by calendar month, whatever dates your statement covers.
For a pivot-style total without building a pivot table, use =SUMIFS($F$2:$F$300,$G$2:$G$300,"2026-04"), where F holds the signed amount and G the month tag. The criterion is text, so the quotes matter.
When does a sheet stop being worth the effort?
If you maintain several cards and accounts, the spreadsheet becomes a chore. SpendixAI accepts PDF, CSV and Excel statements and shows the rows for review before saving, so you can skip most of the repair work and keep the sheet only for odd experiments.
Useful ideas start with a clearer picture of your own spending.
Explore your reports

