Mixpanel

💪 Difficulty: medium-to-slightly-difficult

⏳ Time to complete: approx. 10 minutes

💬 Description

Mixpanel is a event-based product analytics tool that helps product owners, developers and designers understand the way users use their product. We can extract all events going through Mixpanel.

🤑 Use Case

Synthesize how users behave with your product. The data from Mixpanel becomes most interesting in conjunction with other data. For instance, you can A/B test email content (with Mailchimp or Customer.io) and track how the behavior of certain users changes.

🛠️ Setup

Setting up Mixpanel is slightly complex. We outlined all the necessary steps below. If you want us to help you do it, feel free to schedule a call here.

You must be an Admin or Owner in Mixpanel to set up the following

Step #1: Find your project

Select the project from which you want to export data

Step #2: Go to settings

Select “Project Settings” for the particular project

Step #3: ID and Data Residency

Note down your Project ID and your Data Residency.

Step #4: Service Accounts

Go to “Service Accounts”

Step #5: Add new Service Account

Click “Add Service Account”

Step #6: Create Service Account

Input a name for your Service Account, set the project role to “Consumer” and Expiration to “Never”.

IMPORTANT. After clicking “Add” you will be shown a “Username” and “Secret”. You must copy/paste these into a notes program, password manager or whatever your company uses to store password. In other words - don’t just rush through the next steps.

Step #7: Copy/paste username and secret

Copy the “Username” and “Secret” and store them in a secure location. Note that we put a yellow box over our internal credentials below.

Step #8: Input all in Less

Go to your Less platform and input all the 4 parameters in the Mixpannel Connector configuration window. Click “Save Changes” and you’ve set up your Mixpanel Connector.

Step #9: Select number of days to look back

You need to input how many days of data you want to include. If you input 365 then we’ll extract a running 365 days every day. So on Jan 1st, 2000 we’ll extract data from Jan 1st, 1999 - Jan 1st, 2000. The next day, we’ll extract from Jan 2nd, 1999 - Jan 2nd, 2000. You get the point...