CSV Commas Shifting Columns? The Changing Room Bench

2026-11-01

It is 12:40 on a Wednesday. You are sitting on a wet bench in a leisure-centre changing room, hair dripping onto your jumper, and your phone is buzzing in the locker.

You only wanted forty minutes in the pool. It was the one thing you had put in the diary this week, just for you.

You check the phone anyway. It is Jonas, the owner of the small garden-supply shop you do data jobs for. "Import failed again. Half the customers have their postcode in the phone column. Can you look today?"

Your stomach does the small drop you know well.

The row that looked fine

You already know what it is, because it is always something like it.

Somewhere in his export there is a customer whose address field reads Unit 4, Rear of Mill Lane, Hartley. Or a note that says called, will pay Friday, wants delivery before 9. A person typed a few commas into a free-text field, and the file did what files do. It treated each comma as a new column.

From that row downward, everything is off by one. Names sit in the email column. Emails sit in the city column. The order total has quietly slid into a field called "notes".

The cruel part is that the file looks fine at a glance. You scroll through fifty rows and nothing is wrong. The damage is in row 3,118, and you find it only because the import tool throws a red message at you and refuses to continue.

What it quietly costs

Nobody ever puts this on an invoice.

It is the forty minutes you spend hunting for the broken row. It is the second broken row you find after you fixed the first. It is the moment you wonder whether you caught all of them, or whether a customer's phone number is about to end up on the wrong person's record.

It is also the small embarrassment of telling a client, "Give me until this afternoon," for something that sounded like five minutes when he asked.

You dry your hair with the thin towel and think about the lunch you will now eat at your laptop. The swim has already stopped helping.

You are not bad at this. You are good enough at it that people hand you their messy files. That is exactly why it keeps landing on you.

The bench thought

Sitting there with one sock on, you notice something. You have fixed this exact problem before. Different shop, different export, same shifted columns. Last spring it was a café's supplier list. Before that, a newsletter list where someone had put a comma in a company name.

Every time you solved it as if it were new. You opened the file, squinted, patched it by hand or with a one-off script, and then forgot how you did it.

The thought arrives quietly, without drama. It doesn't have to be a fresh emergency every time.

There could be one calm routine. You point it at the file and it shows you what it would change, without touching anything yet. Rows that don't line up get flagged instead of silently swallowed. You get a short report you can read in a minute, so you know what happened and can tell Jonas with a straight face.

That is all you really want. You don't want to be clever about commas. You want to stop being surprised by them.

An ordinary Monday, after

Picture a different Monday morning.

Jonas's export lands in your inbox at 8:15, while the kettle is still ticking. You drop the file into the same routine you always use. A minute later you are reading a plain list: forty-one rows looked odd, here is what was adjusted, here are the three that need a human eye.

You look at those three. One is the Mill Lane address again. You smile a little, because it no longer ruins anything.

You send the cleaned file with a short note. Jonas replies with a thumbs-up and nothing else, which in your line of work is the best review there is.

By 8:40 you are on to the work you actually enjoy. The swim on Wednesday is just a swim. The phone stays in the locker, and nothing in it is waiting to pull you out of the water.

The tools did not become magical. You simply stopped carrying the same small problem around in your head, week after week.

Back on the bench

Back in the changing room, you pull on the second sock and read Jonas's message once more. This time you answer it kindly and calmly: "On it. You'll have it by three."

You know you will probably do it the old, slow way today. But you make a note on your phone, right there with the wet towel on your knee: Stop fixing this by hand.

If any of this sounded like your Wednesday, you might like the Clean Data in Minutes – CSV & Excel Cleaning Toolkit (Python). It helps you clean messy CSV and Excel exports: dedupe, fix dates, phones and encoding, split, merge and validate. It does a dry-run first and gives you a change report, so you can see what would happen before anything is touched. It comes with a 30-day money-back promise. If it doesn't help, you get a full refund, no questions.

No rush. The next messy export will wait for you, and so will the pool.

An illustrative story, written with AI assistance.