Introducing Mirror: Product analytics on your warehouse data

We are thrilled to introduce Mirror, a new mode in our Warehouse Connectors that solves the historic challenge of data distrust due to event immutability.

It’s a well-known issue that product analytics investments often fail due to untrustworthy data. A perfect dashboard is useless if it’s based on inaccurate data. Recognizing this, teams are increasingly pulling trusted data from their warehouse into their analytics tools, a capability we enabled with our Warehouse Connectors earlier last year.

However, our initial version, like typical warehouse integrations, still faced the historic challenge of event immutability. All product analytics tools can append new events, but they can’t modify data that’s already been implemented. This data model challenge inherently limits a warehouse integration. Since warehouse data is constantly changing, such as enriching data or correcting mistakes, event immutability creates the risk of data drift over time.

Mirror marks a significant leap in product analytics because it overcomes this historic data model limitation. Now, Mixpanel data automatically syncs with your warehouse by accurately reflecting all changes, including additions, updates, or deletions.

This perfect sync means that the data you use in Mixpanel is as reliable and current as your warehouse data, while still offering the same power and performance you expect from product analytics. Enabling this functionality required fundamental changes to our data model, pushing product analytics beyond immutable clickstreams into an era of faster, dynamic analysis.

Now you can analyze your trusted warehouse data in Mixpanel, tapping into critical business data like transactional data, revenue, support tickets, and ad spend—the possibilities are endless. As long as an action can be attributed to an identity with a timestamp, it can be modeled as an event and analyzed in Mixpanel.

We’re also expanding Mixpanel’s analytical capabilities to leverage this rich data. While events are often the focus, profiles representing end users or groups are just as important. Every event is linked to a profile, whose properties can change over time. Traditionally, our data model stored the most recent profile value, which is useful for analysis such as tracking signup dates, last seen timestamps, or cohort membership.

With our new Profile History feature, you can now segment and analyze users and groups based on their attributes at any given time. For example, you can understand actions taken by a customer on a free plan before upgrading, or track subscription revenue changes over time.

Mirror and Profile History are available now. Learn more about getting started with them here.