Importing CSV fails for simplest dataset

I’m trying to import my PAXG wallet, using the Import CSV option. But whatever I try to import I get “Something went wrong. Make sure your data is correct and try again”. A UX point first of all - there’s no way to find out if my data is correct at this point, since I can’t examine the spreadsheet once this error occurs. Also, why not allow me to create a manual wallet then start entering data through forms? I’ve only got six rows to add.

I’ve tried importing the initial USD deposit on its own. I’ve tried importing the PAXG purchases on their own. I’ve tried all the transactions together. I’ve logged out and back in again. I’ve tried the interactive importer to manually enter my data. I click the “Only show rows with problems” selector - nothing is shown.

I’m using Chrome on a Mac, I’m on day one of a free trial of the Pro Tier (although I tried it with just the free plan before that with the same result).

An example row, typed into the manual spreadsheet was (with numbers changed):


If I can’t get this data imported I’ll have to walk. I should say that my Celsius CSV import worked fine and my other wallets and exchanges are also playing nicely.

Having the same issue with my BitGo wallets. Please advise.

Hi @robr,

Welcome to the CoinTracker forum. It’s a good point about the UX — we’ll work on improving this. The issue seems to be the date format. It needs to be MM/DD/YYYY HH:MM:SS

Here are the full CSV format specifications. If you have any further problems, please feel free to shoot the CSV to feedback [at]

Hi @amanda,

For BitGo you should be able to upload your unedited BitGo CSV here.

Hi Chandan, thanks for getting back to me. That did the trick, when I further transposed the DD and MM to match the specifications you linked to :slight_smile:

But… it leads to an observation. That date format is pretty US-specific and I’m sure the library you’re using for CSV import can cope with other well-formed dates if you let it. Could you (Cointracker, that is) not allow multiple date formats for the manual import? I’ve just spent a while recasting my manual data (not just the PAXG, but also an ancient set of transactions from that almost slipped through the net) into your format and it’s not a job for a human :). If there’d been any more than 20 rows I would have coded something myself to convert the data.

Anyway, mainly big thanks for the speedy response. Impressive support.

1 Like

Thanks for the feedback @robr. We will work on making this improvement and let you know once it is in place.

1 Like

Which is it? You say it needs to be in DD/MM/YY HH:MM:SS format yet the link you provided to the format ( specifically shows as MM/DD/YY
If you download the example CSV it is also MM/DD/YY.
I’m guessing MM/DD/YY is correct? Is the time in UTC?

Hi @patrick,

Sorry, it should be MM/DD/YYYY HH:MM:SS in UTC

1 Like

While on the CSV subject, should CSV imports always be additive? We can’t just re-import a new full CSV export and have it notice the duplicates?

Hi @patrick,

Not yet — we are working on automatic de-duplication but until then, please make sure not to import the same set of transactions twice.

The csv format is extremely minimal unfortunately. If you had a transaction field for ex then you can ignore duplicates if the transaction field was the same.

1 Like