Indeed UX Personalization

Indeed UX Personalization

Using data-driven design to craft a tailor-made authentication experience for our users

Using data-driven design to craft a tailor-made authentication experience for our users

Using data-driven design to craft a tailor-made authentication experience for our users

  • 01 Problem & Goal ————

  • 03 Solution ————

  • 04 Outcome 🌟 ————

  • 05 Reflection ————

Problem & Goal

Indeed is constantly working to improve the security of its users while maintaining a seamless sign-in experience. This has led to the addition of a wide variety of authentication options that are often presented in a piecemeal way.


This currently causes several issues with the authentication experience:


• Unclear golden path for our users leads to auth delays


• The burden of choice causes unnecessary stress and friction for users


Improper or unusable authentication methods are sometimes displayed


Objective: Create a unified approach to managing and presenting these authentication experiences in a way that is tailored to the individual user based on known traits and past behavior. This gives the user:

1. A clear and simple path to authenticate

2. The best options for them, readily accessible

3. Total control over their authentication experience

“Design is really an act of communication, which means having a deep understanding of the person with whom the designer is communicating.” ~ Don Norman

Research

Research

Mapping the System

I created a comprehensive view of the architecture, interactions, and data flows to ensure a seamless and scalable design. This process requires understanding the user’s journey, aligning business goals, and coordinating with cross-functional teams to identify key touchpoints and dependencies.


Authentication options: Understanding all the available options and how they interact


Entry points: Documenting all of the various entry points into auth flows


Exit strategy: Tracking post-auth experiences so we can ensure a smooth transition

Workshops

Since this initiative impacted nearly every aspect of the Indeed experience, effective cross-team communication was critical to its success. What made this project particularly unique was its dual focus on both employers and job seekers simultaneously.


We scheduled a series of workshops, each tailored to the needs of these two stakeholder groups. These workshops were divided into three focused sessions, each with a clear objective:


Session 1: Understand the problem and identify users’ needs.


Session 2: Present and discuss personalization flows, then vote on which flows to advance to the design and testing stage.


Session 3: Review designs based on the chosen flows and refine them in preparation for user testing.

Following the workshops, rapid prototypes were developed and tested with users to gather actionable feedback.

Overview of a completed workshop. Download the full files below:

Overview of a completed workshop. Download the full files below:

Download the full files below:

Overview of a completed workshop. Download the full files below:

Key Takeaways

Upon completion of the research, I condensed the findings down into a few key takeaways. These would provide the focus for initial and final design iterations:


  • Keep it simple: Users are often overwhelmed by choices during authentication, especially if they are shown irrelevant options.

  • Have a bailout plan: While the data collected gives us a good idea of a user’s preferred auth method, always have a fallback option.

  • Right on time: Utilize progressive disclosure to only show relevant options when the user wants or needs them.

Following the workshops, rapid prototypes were developed and tested with users to gather actionable feedback.

Solution

Using data collected on users—such as their user type, past sign-in behavior, email domain, and verification status—we developed a dynamic authentication system that intelligently predicts their preferred sign-in method. This predicted method is then presented as the default experience each time they return to the site, streamlining the login process.

The solution also prioritizes surfacing safer and simpler authentication methods over defaulting to a password-first experience.

Through user testing, we learned that users appreciate a tailored experience but still need fallback options to avoid frustration. To address this, we designed the system to highlight the most likely authentication method while providing a clear and accessible path to explore all available options if needed.

Initial Release

To mitigate the risk of negative user reactions, we implemented a phased rollout strategy.


The initial phase focused on introducing email OTP as the first authentication method, as it serves as the fallback option for most other methods.


Here you can see an example of one of many possible flows that will be served to a user based on their behavior

To mitigate the risk of negative user reactions, we implemented a phased rollout strategy.


The initial phase focused on introducing email OTP as the first authentication method, as it serves as the fallback option for most other methods.


Here you can see an example of one of many possible flows that will be served to a user based on their behavior

To mitigate the risk of negative user reactions, we implemented a phased rollout strategy.


The initial phase focused on introducing email OTP as the first authentication method, as it serves as the fallback option for most other methods.


Here you can see an example of one of many possible flows that will be served to a user based on their behavior

To mitigate the risk of negative user reactions, we implemented a phased rollout strategy.


The initial phase focused on introducing email OTP as the first authentication method, as it serves as the fallback option for most other methods.


Here you can see an example of one of many possible flows that will be served to a user based on their behavior

Final Designs

Since this initiative impacted nearly every aspect of the Indeed experience, effective cross-team communication was critical to its success. What made this project particularly unique was its dual focus on both employers and job seekers simultaneously.


We scheduled a series of workshops, each tailored to the needs of these two stakeholder groups. These workshops were divided into three focused sessions, each with a clear objective:


Session 1: Understand the problem and identify users’ needs.


Session 2: Present and discuss personalization flows, then vote on which flows to advance to the design and testing stage.


Session 3: Review designs based on the chosen flows and refine them in preparation for user testing.

Following the workshops, rapid prototypes were developed and tested with users to gather actionable feedback.

A comparison of the current and updated UI. Slated for release in Milestone 2.

A comparison of the current and updated UI. Slated for release in Milestone 2.

A comparison of the current and updated UI. Slated for release in Milestone 2.

Outcomes

Milestone #1 was launched and the team collected data as the release progressed.


Results were impressive and showed a significant shift towards the more secure options:

  • New Passwordless Users: 300,000,000+

  • Passwordless conversion: 71.3% to 92.1%

  • Sign-in success rate: >90% compared to 52.1% with passwords

300,000,000+

New passwordless authentication users

300,000,000+

New passwordless authentication users

300,000,000+

New passwordless authentication users

71.3% → 92.1%

Improved passwordless conversion rate

71.3% → 92.1%

Improved passwordless conversion rate

71.3% → 92.1%

Improved passwordless conversion rate

50% → >90%

Sign-in success rate with personalization vs password

50% → >90%

Sign-in success rate with personalization vs password

50% → >90%

Sign-in success rate with personalization vs password

Reflection

Reflection

The Milestone 1 release offered a wealth of insights that informed the strategy for the upcoming Milestone 2. During this phase, we identified several areas where the user experience could be significantly improved:

  • Enterprise level employer accounts will need additional attention to accommodate advanced authentication such as SSO, FIM, and EHS

  • As new authentication methods are developed and implemented - such as passkey - the UI will need to scale

  • Finally, fallback and bailout options, while available, were not always intuitive or prominently displayed, leading to frustration when users encountered edge cases or preferred alternative authentication methods. More refinement can be done here.

These lessons highlighted the need for a flexible, future-proof design. For Milestone 2, the focus shifted to understanding the needs of enterprise level users and other users with advanced or unusual authentication needs. This iterative approach aimed to build a more robust and user-friendly authentication experience.

Additional Information