Boosting Login Success Rates: How Analytics Increased User Onboarding by 29%

Are you tired of seeing potential users drop off during the onboarding process? Imagine turning that frustration into a powerful growth opportunity. This case study explores how a social app developed by a small team leveraged analytics to improve the onboarding user success rate by an impressive 29%.

In this article, we'll walk you through our journey, from the initial hypothesis to uncovering critical insights and implementing impactful changes.

Initial Hypothesis and Challenges

We suspected login and signup processes were creating friction for users. However, lacking concrete data, we couldn't pinpoint the exact issues hindering onboarding. While basic analytics events were implemented, they weren't actively analyzed to identify improvement opportunities.

Defining Success and Setting Up Analytics

To measure onboarding success, we focused on the percentage of first-time users successfully logging in or signing up within a week of installation. Google Analytics and Looker Studio were used to create a dashboard tracking this metric. We analyzed signup and login attempts together, considering them branches of the same onboarding flow within the mobile app (separate measurements might be more relevant for web apps with frequent logins).

Discovering the Problem

Our initial analysis revealed a concerning statistic: only 52% of app installers were successfully onboarded. This meant nearly half of our potential user base was lost during the onboarding process.

Digging Deeper

Beyond the overall success rate, we delved into additional metrics:

• Median time to onboard: 2 minutes 45 seconds
• 90th percentile of time to onboard: 2 hours 54 minutes (indicating 10% of users took over 3 hours to log in/sign up)

We initially considered splitting the onboarding process into signup and login flows for analysis. However, user behavior analysis revealed a surprising trend: 10% of users switched between signup and login attempts more than six times. This fluidity made it difficult to definitively categorize user intent (signup vs. login).

Shifting Focus to User Goals

To gain a clearer picture, we segmented the onboarding process into smaller user goals:

1. App Launch
2. Viewing the First Onboarding Screen
3. Choosing Between Signup or Login
4. Successful Onboarding

For each step, we measured success rates:

• Displaying First Onboarding Screen: 87%
• Choosing Signup or Login: 94%
• Signup/Login Success: 64%
• Overall Onboarding Success: 52% (as mentioned earlier)

A Shocking Discovery

The most surprising insight was that only 87% of new users even saw the first onboarding screen. This translated to a 13% weekly user loss at the very first step.

Understanding the Dropoff

Analyzing the success rates, we identified a significant dropoff between viewing the first onboarding screen (87%) and successfully onboarding (52%). While the signup/login success rate itself (64%) wasn't terrible, the initial screen visibility issue was a critical bottleneck.

Leveraging Analytics for Root Cause Analysis

Since we were using Google Analytics with BigQuery integration, we were able to analyze specific user sessions (without any personally identifiable information) to understand user behavior patterns. Given the large number of failed onboarding attempts (around 1500 users weekly), individual analysis wasn't feasible.

Grouping User Failures:

To identify recurring issues, we implemented the following approach:

1. Select a random user who failed to onboard.
2. Analyze all events from that user's session.
3. Develop an analytics query to isolate the reason for the potential failure.
4. Group all users with similar failure patterns.
5. Repeat steps 1-4 with different random users until common failure reasons emerge.

This process revealed the most frequent roadblocks to successful login:

• Phone Verification Issues: 642 users
• Lack of Login Motivation: 252 users
• Deep Link Problems: 232 users
• Unknown Reasons: 155 users
• Email Verification Issues: 64 users
• External App Login Issues: 54 users

Taking Action

The data clearly pointed towards deep link handling as a major pain point. Years ago, during deep link implementation, a decision was made to display a generic "not logged-in" message upon clicking a deep link while not logged in. This decision, based on the assumption of infrequent occurrences, was hindering user experience.

We also identified and addressed frequent payment issues causing SMS verification problems in certain countries.

The Results

The implemented improvements yielded significant results:

| Metric | Before | After | | --- | --- | --- | | Overall onboarding success ratio | 52% | 67% (improve) | | Success ratio of displaying first onboarding screen | 87% | 97% (improve) | | Success ration of choosing sign-up or log-in option | 94% | 94% (no change) | | Success ration of sign-up or log-in: | 64% | 73% (improve) | | Median time to onboard | 3m 22 seconds | 2m 45 seconds (improve) | | 90th percentile of time to onboard | 2h 54 minutes | 1h 12 minutes (improve) | | Daily Active Users | ~14k | ~19k (improve) |

Investment in Insights: A Winning Strategy

Implementing the analytics solution required a total of six weeks of developer effort, with one week dedicated to development work and four weeks for setting up analytics and building dashboards. While this investment might seem significant upfront, it pales in comparison to the benefits achieved.

Here's why this approach proved far more successful than blindly guessing at improvements:

• Data-Driven Decisions: Previously, improvement efforts were based on hypotheses, lacking concrete data to pinpoint the root causes. This could have led to changes that offered minimal improvement or even negatively impacted the app.
• Targeted Solutions: The analytics identified specific user pain points, allowing us to address the exact issues hindering onboarding. This resulted in a much more impactful improvement (29% increase in onboarding success rate).

Looking Ahead: Unlocking Further Growth

While we've made significant strides, there's always room for improvement. The "I'm not convinced to log-in" group identified through user session analysis is a prime example. We can use this insight to explore strategies like in-app tutorials or app feature previews to incentivize login and improve user activation.

Our success with onboarding analytics highlights the power of data-driven decision-making. By continuing to leverage analytics throughout the development process, we'll be well-positioned to tackle future challenges and unlock further growth for the app.

Summary

Our data analysis increased user onboarding success by 29% in a social app. Initially, we identified issues with login and signup processes but lacked concrete data. Using Google Analytics and Looker Studio, we tracked and analyzed onboarding, revealing that only 52% of new users completed the process. By segmenting and analyzing user behavior, we identified key obstacles such as phone verification and deep link issues. Addressing these led to a 67% onboarding success rate and an increase in daily active users from 14k to 19k. This case highlights the value of data-driven decisions in improving onboarding and driving app growth, which can be implemented even by a small team.

Ready to Achieve Similar Results for Your Company?

We specialize in delivering high-impact solutions with small, agile teams. If you're facing challenges with your app's performance or user experience, we can help you achieve similar outstanding results. Contact us to see how we can tailor our expertise to meet your company's unique needs and drive your success.