CSV export seems weird?


Userlevel 1
Badge

Hello

Exporting CSV files seems to be acting weird. The exported file seems to be seperated with the “,” sequence of character quotation marks included, instead of just the , character

 

This has a result neither the Sprig platform nor excel to be able to read them properly. Any ideas? 

 


13 replies

Userlevel 7
Badge +10

Hey @stelios 

I tried exporting a few retention chart CSVs on my end to see what’s happening and if at all the number of segments made any difference.

It looks like the badly formatted output is for charts where the time range is selected using Since and Last X days. If I export the same chart, but by using the Between Date A and Date B selection the exported CSV is formatted correctly. Can you try verifying this behavior on your end?

Not sure if this is a bug or the expected output behavior, but looping in @Amplitude Admin to look into this.

Hope this helps.

 

Userlevel 6
Badge +9

Thank you for looking into this, Saish. Our platform specialist @ning.chang will be able to help. 

Userlevel 6
Badge +8

HI @stelios ,

 

I am not aware of any issues with the CSVs output recently. Would you be so kind as to send me the link to your chart? (I will be able to access it) and let me know what you expected the output to be? I will then download the CSV and see if I can replicate the behaviour. Something that might help is to get the CSV and open it in Microsoft Excel, click the A column entirely, go to the data tab, do Text To Columns and choose Delimited, go next and then choose Tab, Semicolon and Comma and click finish. This should present the excel in a better format too. I hope this helps and look forward to your reply!

 

Kind Regards,
Denis

Userlevel 1
Badge

https://analytics.amplitude.com/fanfest/chart/5osfha5/edit/jndqsy7

 

Here. I am exporting the breakdown as a CSV. 

Userlevel 6
Badge +9

Thank you @Denis Holmes ! :grinning:

Userlevel 1
Badge

Hey I did what you said and it seems the first days are “dominated” with the users, it still does not appear correctly
 

 

Again your CSV in a text editor looks pretty weird with some peculiar tab characters (It seems over engineered and bloated). I tried to import CSV cohorts to Sprig (https://sprig.com/) and it couldn’t read your CSVs and while chatting with their support they told me that your CSVs are a known issue. 

 

Userlevel 1
Badge

Right I figured out part of the issue: 

tThe weird lines are because I am excluding some emails from my research. 

 The exclusions, for whaterver reason are being passed down as data in the CSV. It’s these weird things here:

If I remove the exclusion it looks better in Excel, but Sprig still cannot read it:

 

 

With all good intentions some feedback. This is a good CSV:

 

This is a bloated, over engineered mess: 

 

Tabbed characters, Quotations , a big title up , with links up there and what not. 

 

I apologise for the harsh criticism, but right now I am trying to get my job done and I cannot. I am talking with a poor guy from Sprig’s support (by the way why don’t you have chat support?)  and he is trying to clean your CSV so that it can get uploaded. Something that was supposed to be your job in the first place and not theirs. 

Quite disappointed :no_mouth:

 

Userlevel 6
Badge +8

Hi @stelios ,

 

I appreciate the feedback and I understand the frustration. Can you please PM me your email address you use in Amplitude? I can then start a  ticket for this instead as it will require further investigations. 

 

Our chat support is only for paying customers unfortunately, we use Community for our Starter customers. If you can PM me your email address you use in Amplitude, I will begin the ticket. Thanks!

 

Kind Regards,
Denis

Userlevel 6
Badge +8

@stelios Can you also try this workaround provided by a colleague and see if that works? I know it is not ideal but I am hoping this will help us along ! Can you Open the CSV in Google Sheets and then save it as an excel? WHen I open it in Google Sheets, I get the below ;

Can you see if this helps? Thank you!

Userlevel 1
Badge

Did that and didn’t work. Sprig still doesn’t read it. My solution was to open it in Sublime and delete all the egregious tabular characters, whitespaces, “:” characters and whatnot. It was a mess. Then it read it properly. 

It’s a simple CSV file, they have existed as a file format since 1972, it’s data with commas. That’s all: Data with commas. This is what happens when someone tries to be smart with solved problem. Tell your product person that CSV files don’t need links, or headers or whatever. It’s not the Mona Lisa it’s a CSV file. 

Thanks for the suggestion though… Honestly.

 

Userlevel 6
Badge +8

@stelios ,

 

I will pass on your feedback to the product team. I hear you loud and clear. I am glad you got it working albeit the hurdles. I do understand and I appreciate the really detailed feedback. I am going to pass this onto the appropriate feedback team for their consideration.

 

If there are remaining issues, do send me a PM with your email address and I will be happy to investigate on a ticket. 

 

Kind Regards,
Denis

Hi Denis, 

Chiming in here: as a marketer, I love .CSV files, and within 3 hours of getting access to Amplitude, I came across the same issue as @stelios.  I have a small export of 70 users that I want to upload some of their data into Marketo, but I can’t since there’s quotation marks and 7 spaces at the beginning of each cell’s data. 😓

 

I do hope that this product feature is updated soon so the tool is more accessible to those working alongside product teams. 

 

Thanks, 

Tiffany 

Userlevel 6
Badge +9

Thank you @tmosco23 😀 We appreciate your involvement in the Amplitude Community. I’ve also shared this feedback with the product team, who will be able to take a close look! 

 

Reply