Having issues bulk inviting a CSV of users

[Camille_Augereau]

@techAPJ

So you want to invite users to your Discourse instance, but you have so many that sending invites individually would be a lot of work? Good news – you can send bulk invites! Here’s how:

Prepare CSV File

Prepare a CSV file with the users you want to invite. These points should be considered while preparing CSV file:

• One user per line.
• Email is required in the first column, and the email must be valid.
• Any permissions groups you want this user to be a member of should be in second column. For multiple groups, separate group names with a semicolon group_1;group_2;group_3
• Normally invited users arrive at the homepage. If you would rather invited users end up on a specific topic, enter that Topic ID in the third column.

The format is:

name@example.com,group_1;group_2,topic_id

Note that group names and topic id are completely optional, only email is mandatory. You can have an invite that is just the email alone:

name@example.com

It’s easy to create CSV files in a spreadsheet, which would look like this:

Thank you for this explanation but I tried to do a mass invitation and it didn’t work, my file is in CSV but all my emails are written like invalid email, I don’t understand why

[simon]

@Camille_Augereau

my file is in CSV but all my emails are written like invalid email

What application are you using to create the CSV file? Is the message you receive from Discourse after uploading the file telling you that the emails were invalid?

[patrickemin]

Same here, never got this to work.

[simon]

What application are you using to create the CSV file?

[patrickemin]

Excel (Office 365 suite)

[simon]

@patrickemin

Excel

I would be surprised if that is causing issues with the CSV file. Do you get an error message on Discourse when you upload the CSV file?

[patrickemin]

Hello, yes, this is the error I get:

[codinghorror (Jeff Atwood)]

We’ve seen this before when the CSV has a weird encoding or is saved improperly. Are you 100% sure it’s plain vanilla, plain text CSV with no strange encoding?