Mixpanel helps you study user behavior to make better decisions about your product. This article provides a high-level overview of the core concepts of Mixpanel and the analyses you can do.
Types of Analysis
Most analysis about user behavior in products boils down to 3 types:
- How many rides were completed in the US in the last 30 days?
- How many users did a purchase yesterday?
- How many minutes of content was viewed across all shows in India?
- How many new users do I have each week?
- You can answer these questions in the Insights report.
- What was the conversion rate from add to cart to checkout yesterday?
- How many users listened to 5 songs and created a playlist within the same session yesterday?
- How quickly do users convert from Signup to Purchase?
- You can answer these questions in the Funnels report.
- What was the 7-day retention for users sending text messages?
- How many users come back within 6 months of signing up?
- You can answer these questions in the retention report.
- Event: An event is a data point that represents an interaction between a user and your product. Events have a name ("Purchase"), a timestamp ("2021-03-01 12:30:00"), a user ID ("email@example.com") and a set of properties ("item": "iPhone", "price": 700). Events are generic enough to describe any action that a user takes at some point in time.
- User: A user is an individual that interacts with your product. In Mixpanel, each event is performed by a user.
Property: Named piece of context attached to an event or user. There are 2 types of properties:
Profile: A profile is a collection of profile properties that describe the most recent context about a user, group or another property. There are 3 types of profiles:
- User profile: Collection of profile properties that store the most recent context of a user.
- Group profile: Collection of profile properties that store the most recent context of a group.
- Lookup profile: Collection of profile properties that store the most recent context of another property. Let's say you are storing SongID as an event property, but now want to expand SongID to also store Song Name and Song Genre, a lookup profile property on SongID can let you do that. To read more, view Lookup tables.
- See this to understand the differences between a user profile, group profile and a lookup profile.
- Metric: Metrics summarize events, users, or properties into a number.
- E.g. Video events -> Total videos watched in last 30 days.
- Segment: are subsets of events or users that share a set of properties.
- E.g. Users that watched a video from city = San Francisco.
Breakdown is another word for segmentation. To breakdown by a property is to split events or users into segments, each of which shares a particular value, or range for that property.
Breakdowns let you view multiple segments. Filters just zoom in to a particular segment.
Time is a property on each event that indicates when it happened.
- Selecting a range in the datepicker scopes down analysis to only events during that time range.
- A line chart is how we express segmenting on time. Every point on the chart is based on the events from that time bucket.
A virtual event that was created from one or more events, optionally filtered down by a particular set of properties, and given a name.
- Merge two events into a single event ("User Signup" + "Account Created" -> "Signup")
- Create an event based on a filter on another event ("Purchase where Country = "US" -> "US Purchases")
Properties created from other existing properties by using Excel-like formulas. Learn more.
A cohort is a segment of users that shares a particular property or a behavior (event count, property sum, distinct days active, etc.)
Reports are interactive ways to define, visualize and segment metrics in real-time.
- Insights: Insights lets you visualize a metric (count, sum, average of properties...) of your events or users, optionally filtered or segmented based on their properties. Learn more.
- Funnels: A funnel lets you measure user conversion across a series of events. Learn more.
- Flows: Flows identifies the most frequent paths taken by users from or to any event. Use Flows to understand how your users sequentially perform events in your product, and analyze drop-offs or unsuccessful behavior. Learn more.
- Retention: Retention measures how different cohorts of users stay engaged over time.
- Users: Users shows you a list of your users, optionally filtered down to some segment with a shared property or behavior. Learn more.
A dashboard is a collection of reports. A dashboard can have a filter applied. In that case, the filter is applied to all of its constituent reports.
Compare to Past
Compare to past is a shortcut for segmenting a visualization on two time periods (current and past).
This is a period of continuous user activity. Use Sessions in Mixpanel to track which events users perform within a session, how many sessions contained a key action or how many sessions it takes for a user to convert in a funnel. Learn more.