Conversion Optimization
Conversion optimization helps you understand where users complete key actions in your product, where they drop off, and what drives them to convert. Mixpanel’s Funnels report is designed specifically for this type of analysis.
Understanding Funnels
Use the Funnels report to analyze where your users convert, how long they take to do so, and what drives conversions.
A funnel is a sequence of events that represents a critical user journey in your product. Examples include:
- Signup → Email Verified → Profile Completed → First Action Taken
- Product Viewed → Added to Cart → Checkout Started → Purchase Completed
- Free Trial Started → Feature Used → Plan Selected → Payment Added
Configuring Conversion Criteria
To utilize funnels correctly, it’s important to understand how the conversion criteria works and how it impacts the conversion rate depending on whether you use “unique” or “total” conversions.
Unique vs. Total Conversions
Unique conversions count each user only once, even if they complete the funnel multiple times. This is the most common setting and gives you a clean view of how many individual users convert.
Total conversions count every instance of funnel completion. Use this when you want to measure repeat behavior (e.g., how many purchases happen, not just how many users purchase).
For most product funnels (signup, onboarding, activation), use unique conversions. For transactional funnels (checkout, booking), total conversions may be more relevant.
Analyzing Your Funnel
Once you’ve built your funnel, look for these key insights:
Overall Conversion Rate
What percentage of users who start the funnel complete all steps? This is your baseline metric.
Step-by-Step Drop-Off
Where do users drop off? The step with the largest percentage drop is often your biggest opportunity for improvement.
Time to Convert
How long does it take users to move through the funnel? Understanding timing helps you:
- Set realistic expectations for conversion windows
- Identify steps where users get stuck
- Optimize re-engagement campaigns
Add 3-5 events that represent your critical user journey.
Identify which step has the largest drop-off.
Break down by user properties (device, source, plan type) to find patterns.
Use Flows or Session Replay to understand why users drop off at that step.
Advanced Funnel Features
Go beyond looking at overall conversion with these advanced capabilities:
Conversion Trend Over Time
View how your conversion rate changes over days, weeks, or months. This helps you:
- Spot the impact of product changes or marketing campaigns
- Identify seasonal patterns
- Track progress toward conversion goals
Time to Convert Distribution
See the distribution of time it takes users to convert through the funnel. This reveals:
- Whether users convert immediately or over multiple sessions
- Optimal windows for re-engagement messages
- Differences between fast and slow converters
Frequency Analysis
Understand the number of times users complete any step before converting or dropping off. This shows:
- How much exploration happens before conversion
- Whether users need multiple attempts to complete steps
- Which steps require the most retries
Hold Property Constant
Add more specificity to your conversion criteria by holding a property constant over steps. This ensures users see the same value for a property throughout the funnel.
Example use case: Track users who view the same product and eventually purchase that specific product, filtering out users who switched products mid-funnel.
Exclusion Steps
Exclude users who did a particular event between funnel steps. This helps you:
- Filter out users who encountered errors
- Focus on the “happy path” experience
- Isolate specific user segments
Example use case: Measure checkout conversion while excluding users who encountered a payment error.
Viewing as a Flow
Oftentimes, knowing what the conversion is or how it has changed over time is not actionable in itself. Beyond creating cohorts on users who have dropped off to send them downstream to CRM tools for re-engagement, what marketers and product managers typically want to know is:
- Where precisely did people drop off?
- What are the typical user flows or behavior that causes users to drop off?
The View as Flow feature helps you answer these questions by visualizing the paths users take after dropping off from your funnel.
Use View as Flow to discover unexpected user journeys that explain drop-off behavior. You may find that users who drop off are actually taking a different path to value.
Segmenting Your Funnel Analysis
Breaking down your funnel by user properties or behaviors reveals which segments convert best:
Common Segmentation Dimensions
- User properties: Device type, location, acquisition source, plan type
- Behavioral properties: Number of sessions, time since signup, feature usage
- Temporal properties: Day of week, time of day, cohort
Questions to Answer with Segmentation
- Do mobile users convert at a different rate than desktop users?
- Which marketing channels drive the highest-quality conversions?
- Are new users more likely to convert than returning users?
- Does conversion rate vary by geography or language?
Click “Add breakdown” and select a property.
Look for significant differences in conversion rates.
If one segment converts much better or worse, dig into why.
Optimize the experience for low-converting segments or double down on high converters.
Creating Cohorts from Funnels
Once you’ve identified users who drop off at specific steps, create cohorts to:
- Send re-engagement campaigns via your CRM or marketing tools
- Analyze their behavior in other reports
- Compare them to users who converted
How to create a cohort:
- In your Funnel report, click on a specific step
- Select “Create Cohort” from the dropdown
- Choose whether you want users who completed or did not complete that step
- Save and name your cohort
Create cohorts of drop-off users to send targeted re-engagement campaigns, or cohorts of converters to identify patterns in their behavior.
Optimizing Your Funnel
Once you understand where and why users drop off, take action:
Reduce Friction
- Simplify forms and reduce required fields
- Improve page load times
- Clarify calls-to-action
- Fix bugs or errors that block progress
Add Guidance
- Provide inline help or tooltips
- Show progress indicators
- Add social proof or testimonials
- Offer live chat or support at critical steps
Experiment with Changes
Use Mixpanel Experiments to test funnel optimizations:
- A/B test different copy or designs
- Test shorter vs. longer flows
- Experiment with incentives or urgency
Monitor Impact
After making changes:
- Set up Alerts on funnel conversion rates
- Use Annotations to mark when changes went live
- Create a Board to track funnel performance over time
Real-World Example
Scenario: An e-commerce site has a checkout funnel with these steps:
- Add to Cart (100% baseline)
- Enter Shipping Info (65% of users)
- Enter Payment Info (45% of users)
- Complete Purchase (38% of users)
Analysis: The biggest drop-off is between steps 1 and 2 (35% of users).
Investigation using Flows: Users who drop off at step 2 are often clicking “Save for Later” or returning to browse more products.
Action: Add a “Save Cart” feature that lets users continue shopping without losing their cart, and send an email reminder 24 hours later.
Result: Conversion from step 1 to step 2 increases to 72%, and overall funnel conversion improves from 38% to 46%.
Key Takeaways
- Funnels help you understand where users convert and where they drop off
- Configure conversion criteria correctly to measure unique or total conversions
- Use advanced features like time to convert, hold property constant, and exclusion steps for deeper analysis
- Segment your funnel by user properties to find patterns
- View as Flow to understand where drop-off users go next
- Create cohorts to re-engage users or analyze their behavior
- Test optimizations with experiments and monitor impact with alerts
Next Steps
- Build your first funnel for a critical user journey
- Identify your biggest drop-off point and investigate with Flows
- Create a cohort of users who drop off and analyze their behavior
- Set up an experiment to test a funnel optimization
- Review the Retention Analysis guide to understand long-term engagement