User Theme Preferences: Analyzing Application Theme Switches | Mixpanel Community

User Theme Preferences: Analyzing Application Theme Switches

kapa.ai I would like to see the total number or percent of users that use my application that have picked one theme over another. There is a default theme that everyone has, but you have the choice to switch it, and switch back as many times as you want. Cohort analysis isn't enough, because it doesn't allow me to look at the most 'recent' value that someone has chosen. However, an Insight report isn't enough either because you have to trigger an event to see that a user made a switch at all.

6 comments

kapa.ai
APP
·
·
Please stand by Iva Đ. as I search the knowledge base. I normally take about 15 seconds or so. 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
kapa.ai
APP
·
·
Based on your question about tracking theme preferences in your application, I can suggest a few approaches using Mixpanel's capabilities.You're right that this is a nuanced analysis case - you need to track both the event of switching themes and the most recent theme selection for each user. Here's how you could approach this:
Using Custom Properties
Mixpanel's Custom Properties feature would be ideal for this scenario. You can create a custom property that borrows the theme selection from the most recent theme-switching event.As described in the documentation, borrowed properties allow you to "borrow properties from one event onto subsequent events," which is perfect for your use case where users can switch themes multiple times: 1. First, create a borrowed property that captures the theme selection from your theme-switching event 2. Then use this property in your analysisLearn more about borrowed properties
Alternative Approach: User Profiles
Another approach would be to update a user profile property each time a user switches themes. This would always reflect their most recent selection, and you could then use this property in your analysis.
Creating the Analysis
Once you have the theme data available as a property, you can: 1. Use Insights to create a breakdown report showing the distribution of users by their current theme selection 2. For the measurement, select "Unique Users" to see the count or percentage of users per theme 3. You can visualize this as a pie chart or bar chart to clearly see the distribution
Behavioral Analysis
If you want to go deeper, you could also use Mixpanel's behavioral properties to analyze: - How frequently users switch themes - Whether certain themes correlate with higher engagement - If users who customize their theme retain betterFor example, you could create a report similar to what's described in the retention analytics primer, where they compared retention rates between users who customize settings versus those who don't: Retention analytics primerThis approach would allow you to not only see the current distribution of theme preferences but also understand how theme selection impacts user behavior and retention.I hope this helps with your analysis! Let me know if you need more specific guidance on setting up any of these approaches.
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
Iva Đ.
·
·
thank you!
Iva Đ.
·
·
kapa.ai actually the borrowed properties provided inconsistent data that severely mismatches what we already have. We do not want to modify our user properties so that every single design choice is a user property. Please provide me with a different solution that is more robust
kapa.ai
APP
·
·
Let me check the knowledge base for you again. One moment! 📚
kapa.ai
APP
·
·
I understand that borrowed properties didn't provide consistent results for your theme tracking needs, and you don't want to overload your user properties with design choices. Let's explore more robust alternatives based on Mixpanel's capabilities.
Historical Profile Values
A more robust approach would be to use Mixpanel's Historical Profile Values. This feature is designed specifically for properties that change over time (like theme preferences): > "Historical profiles layer on additional capabilities by capturing changes in each property over time instead of just the latest value."With historical properties, when you apply a breakdown, "the property value will be picked based on the time of the event, instead of the current property value." This means you can accurately track how theme preferences change over time.Importing Historical Profile Values
Warehouse Connectors
If you have your theme selection data stored in your data warehouse, you could use Mixpanel's Warehouse Connectors to import this data directly: > "Point Mixpanel at a table (or view!), tell us what the timestamp column is, and it'll appear as an event with all its columns as properties. No more ETL jobs, pipelines, or scripts to maintain."This approach allows you to leverage your existing data structure without modifying your tracking implementation.A leap forward in the product analytics data model
Event-Based Approach with Custom Properties
Instead of borrowed properties, you could create a custom property that uses mathematical or logical functions to determine the most recent theme selection. For example, you could: 1. Track theme changes as events with properties indicating the selected theme 2. Create a custom property that uses date/time functions to identify the most recent selection 3. Use this custom property in your analysisThe custom properties feature allows you to "combine existing properties into new properties on the fly, using a simple Excel-like formula language."Custom PropertiesThese approaches should provide more robust solutions for tracking theme preferences without overloading your user properties. Would you like me to elaborate on any of these options?
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.

6 comments

Using Custom Properties

Alternative Approach: User Profiles

Creating the Analysis

Behavioral Analysis

Historical Profile Values

Warehouse Connectors

Event-Based Approach with Custom Properties