How to determine category for Amazon?

what’s the best way to determine category for Amazon transactions? Many of the descriptions show something like: Amzn Mktp Us*wv09w66n3 I can’t tell what the purchase was. And even the dollar amount on Amazon doesn’t match. I’m sure it’s because certain items shipped together charged together and it’s not necessarily how you ordered them. Any tips on easily figuring this out?

I tried using this: How to use the Amazon Line Item Importer for Google Sheets but finally decided a Wild Ass Guess was good enough for my purposes. However it might work for you.

1 Like

As @susandennis notes, the Amazon Line Item Importer is designed to solve the problem you raise, @wendyb5087. I hope you’ll consider giving it a try.

I’m curious to hear more about your experience with the Amazon workflow, @susandennis. Did it not work for you?



It worked but it was a lot of work. And I was afraid of screwing up my spreadsheet. I decided that I could more easily and safely do it manually. PLUS, 85% of my Amazon purchases fall into two categories so it’s not that difficult.

I recommend creating a separate category for Amazon purchases altogether. I personally categorize Amazon as “Miscellaneous” for any transaction under ~$80. Anything over $80 gets looked at and assigned manually as needed e.g. if I buy dining room chairs for $200 I will re-assign them to Home Improvement category etc. The only other stores that get categorized as “Miscellaneous” in my budget are Target and Walmart. I rarely shop there and if I do, it’s usually something random such as baby toys, household items etc that I don’t have a separate category for.
I find Amazon line item importer to be an overkill but it all depends what you buy and how often.

Thanks for the feedback, @susandennis. When you say it was a lot of work, do you mean going through so many transactions one at a time? Or the process of downloading and running the CSV through the processor?

Was part of the issue that made it time consuming that you were trying to go back to the start of your Amazon purchase history?

There is a filter checkbox that you can use to just show the rows that need categorization after an import. My hope was that this workflow would make it easy to categorize the rows you just imported in the Transactions sheet.

Just asking so we can make the tool better.
Thanks for your thoughts.

Personally, I like to be more exacting in my categorization— hard to budget for “Miscellaneous” :wink:— and I also don’t love the workflow of bouncing between the Transactions sheet and my Amazon order history to remember what (even a subset of) the purchases were.

For many people without my personal hangups though I think your approach is a good one, @Exo3.

Thanks for sharing,

Really good questions, Randy. It was back when I first started using Tiller so I was just twitchy about screwing my spreadsheet up. Now that I have, several times and survived :), I should go back and try it again. I will and will let you know, precisely where my pain points are, if any.

If you happen to give it another shot, I’d appreciate your feedback.
Thanks, @susandennis.

P.S. Moments ago, I published a Tiller Labs add-on build with the (Amazon & other data source) CSV import workflows migrated from the Amazon Line Items add-on. You can find the new functionality under the Tools menu. We will obsolete the Amazon Line Item add-on in the coming weeks. If you are up for giving the tool a spin in its new home, you’d probably be the first!

Ok, I gave it another shot. MUCH better this time I appreciated the help info and explanations but still kind of hit a snag with the ‘transfer’ ‘amazon’ option. I didn’t know which one to pick.

I also imported the refunds. Intuitive.

I see how it’s working but still managed to create duplicate expenses in my sheet. Total user error and likely not enough coffee. ALSO mixed with a heavy dose of probably easier for me to do it manually. Really. I have only 1 or 2 of these entries daily. And I check/update my Tiller sheet every morning.

I love and appreciate the options, however!

Good question, @susandennis, regarding the choice between categorizing Amazon orders as Transfer or Amazon

Using Amazon uniquely and intuitively buckets the credit card charges in a way that is easily searchable. Using Transfer aligns intuitively with the Tiller “Transfer” concept (i.e. offsetting charges that net to zero) and cleanly removes the need for an extra category in your sheet (i.e. “Amazon”) that really needs no budget.

They are both valid approaches.

Personally, we had been categorizing as Amazon and, with the start of 2020, decided to change to Transfer. The reason I chose to make this change is that our family is using the line-item/CSV importer to pull data from Amazon, Paypal and Venmo… and I didn’t like the idea of creating a category for all of them— especially when it is really just an unbudgeted staging area until the line-item offset is pulled in. Switching to Transfer felt leaner and more efficient.


1 Like

Thanks so much Randy. Working on it.


Today I tanked my spreadsheet and so then started over from scratch. Not the crisis it sounds like, I’m only interested in 2020 so less than 6 weeks of data. ANYWAY… with a clean slate and a clearer understand of both Tiller AND the Amazon CVS importer, I tried again.

This time, it worked FLAWLESSLY!! Seriously. It’s great. One reason, btw, that it screwed up before was that I did NOT tag everything as Amazon and, yet, selected Amazon instead of transfer so I had duplications and, honestly a mess.

This time, I did it all correctly from the start and it was great. Thank you. S.

Glad to hear it, @susandennis. I’m pretty excited about the importer and use it regularly… so it makes me happy to hear others appreciate it as well.

Honestly, I’ve been tagging both credit card transactions & Amazon Orders as “Transfer” (starting in 2020) since it feels cleaner to me… but it works either way.


I tell you the one snag that still lingers and that’s on Amazon’s end. For several months my CSV of purchases turned up empty every time. Finally, one day, it was not. Now my returns is coming up blank when I know there are at least 7 in the time frame requested.

FYI, when there is no data, your importer’s message is that it is not a CSV. It is, it’s just empty. That took me a couple of rounds to figure out :slight_smile: