Inside Mixpanel: The OADA Loop

A SaaS team observes where new users drop off during onboarding, analyzes behavior to uncover friction points, decides which steps to simplify, and acts by testing improvements that drive activation and conversion.

Observe
The team tracks the onboarding journey to understand where users get stuck: Account Created → Tutorial Completed → Key Action Taken

Using Funnels and Flows in Mixpanel, they see that a large portion of users drop off before finishing the tutorial.

Analyze
They dig deeper with Insights and Session Replay to find out why. Behavioral data shows that most users abandon onboarding when asked to configure a complex setting too early, while replays confirm that this screen causes confusion.

Decide
The team compares cohorts of users who complete onboarding in their first session versus those who don’t. They find that users who finish onboarding right away are twice as likely to activate within a week. They decide to simplify the initial setup and move optional configuration to a later stage.

Act
They run an Experiment testing a new guided flow that delays complex setup until after the user experiences initial value. After the winning version significantly improves activation, they roll it out to all users using Feature Flags.

✨ Result: Activation improves 15%, and time to value shortens.

An eCommerce team observes how shoppers move through checkout, analyzes drop-offs to find mobile pain points, decides what changes to prioritize, and acts by launching experiments that lift conversion rates.

Observe
The team maps the path to purchase to pinpoint where customers abandon the flow: Product Viewed → Added to Cart → Checkout Started → Purchase Completed

Using Funnels in Mixpanel, they see that a large portion of users start checkout but never complete payment.

Analyze
They investigate further with Insights, segmenting results by device and geography. Mobile users show significantly lower completion rates.

Through Session Replay and Heatmaps, they observe users zooming in on payment fields, misclicking form inputs, and abandoning the process after multiple failed attempts.

Decide
The team compares cohorts of mobile users who completed checkout versus those who didn’t. They find that reducing the number of required fields correlates with higher conversion. They decide to simplify the payment form and surface popular payment options earlier in the flow.

Act
They create an Experiment testing the simplified checkout form with mobile users. The results show a significant lift in mobile conversions, so the team rolls out the new design to all users using Feature Flags.

✨ Result: Checkout completion increases by 20%, and mobile users complete purchases faster with fewer errors.

A streaming platform observes how audiences engage with content, analyzes where viewers disengage, decides how to personalize recommendations, and acts by improving the experience to keep viewers watching longer.

Observe
The team maps the typical viewing journey to understand where viewers lose interest: Episode Started → Episode Completed → Next Episode Started

Using Funnels, they see a major drop-off after Episode 2 in new series. Flows confirm that most users who stop watching don’t move on to similar shows or genres.

Analyze
They dig deeper using Insights, Retention, and Session Replay to uncover why engagement drops off. They discover that episodes with weaker completion rates also have fewer post-watch interactions, like “Add to Watch List” or “Rate Content.” Session Replays and Heatmaps reveal that end-of-episode recommendations are often irrelevant or missed entirely.

Decide
The team compares cohorts of viewers who continue past Episode 2 versus those who stop. They find that engagement with “Next Episode” or “Rate Content” interactions strongly predicts long-term retention. They decide to improve personalization and prompt users to rate episodes before recommending what to watch next.

Act
They run an Experiment testing a new “Rate this Episode” prompt and smarter “Up Next” recommendations. Once they see a significant increase in episode-to-episode continuation, they roll out the new experience globally using Feature Flags.

✨ Result: Repeat viewership increases 25%, and average session length grows as more viewers continue past early episodes.

A fintech product team observes how customers engage with budgeting tools, analyzes where they drop off, decides which steps to simplify, and acts by testing new prompts that improve adoption and retention.

Observe
The team maps the customer journey to identify where engagement drops: Account Linked → Budget Created → Spending Reviewed → Budget Adjusted

Using Funnels and Flows in Mixpanel, they see that many users link their bank account but never complete a first budget. This gap represents a key opportunity to improve activation within the product.

Analyze
They dig deeper using Insights, Retention, and Session Replay to understand why users don’t finish setup. Behavioral data reveals that users linking smaller financial institutions often encounter errors. Retention reports show that users who successfully create a budget are twice as likely to return within 30 days.

Decide
The team compares cohorts of users who complete their first budget versus those who don’t. They find that users who set up a budget alert immediately after creation retain best over time. They decide to streamline the account linking flow and prompt users to set alerts earlier.

Act
They launch an Experiment testing a simplified linking process and a new “Set Your First Alert” prompt after budget creation. When the new experience increases completion and retention, they roll it out to all users using Feature Flags.

✨ Result: Budget feature adoption increases 30%, and retention improves 15% as users connect accounts and set alerts more easily.

A gaming studio observes how players progress through levels, analyzes behavior to uncover friction and missed opportunities, decides what changes will keep players engaged, and acts by testing prompts that increase completion and purchases.

Observe
The team tracks key in-game milestones to identify where players churn: Level Started → Level Completed → In-App Purchase Made

Using Funnels and Flows in Mixpanel, they find that a large portion of users drop off after failing Level 3 multiple times. This stage becomes their primary focus for improvement.

Analyze
They use Insights, Session Replay, and Heatmaps to understand what’s causing frustration. Behavioral data shows that players who use “Power-Ups” early in gameplay are far more likely to progress, while replays reveal many players overlook the in-game hint icon.

Decide
The team compares cohorts of players who used Power-Ups before Level 3 with those who didn’t. They confirm that introducing Power-Ups earlier increases both level completion and purchase likelihood. They decide to highlight Power-Ups at the start of Level 3 with a short tutorial prompt.

Act
They run an Experiment testing the new Power-Up prompt, measuring its effect on both level completion and in-app purchases. After the new version shows a significant improvement in completion and a rise in purchases, the team deploys it to all players using Feature Flags.

✨ Result: Players advance further through the game, spend more time in-app, and make more purchases, driving sustained engagement.