Postman: Notification Center 2.0

Brief

Staying up to date on everything related to your project and team should be intuitive and easy. This growth experiment looked at increasing app usage by improving collaboration and awareness of changes within a Postman team.

This work would add:

  • Improve notification alert system
  • Improve quality of notifications, building stronger templates and prioritizations
  • Build stronger usage of in-app systems over email

My Role

As Lead Growth Product Designer, my responsibilities were:

  • Helped established business goals and mapped to user/customer needs
  • Discovered additional opportunities to achieve business goals
  • Mapped the user flow and designed wire-frames
  • Creating high-fidelity mocks and prototype flows
  • Assisted data analyst team with quantitative tracking

The team members

  • Product Designer (Me)
  • 2 Project Managers
  • 3 Engineers
  • 1 Data Analyst
  • 1 UX writer

Journey

Using data to understand the problem space

The value of a Postman team is centralized collaboration within the platform. Users engage with email notifications over in-app notifications. However when diving deeper into the data I noticed some key issues:

  • 0.8% users engage with the in-app notification center week over week (WoW)
  • Users who engaged with notifications were 30% more likely to convert / retain
  • 5x users would disable email over in-app
  • Less than 50% team users are active week over week.

Qualitative dive into the problem

Ran a Heuristic Evaluation and user surveying to understand the pain points user face when trying to collaborate. What I found was:

  • Invites get lost in email
  • Users can’t find recent high priority notifications
  • Often users didn’t realize there was an in-app notification center
  • Notification center is a flat unprioritized list
  • Notifications were visually inconsistent

Market Analysis

I investigated 10+ products to better understand how other systems handle notification and user alert systems. During my investigation I found these factors to be key in how to better inform users:

  • Notifications were are often grouped by value/priorities
  • ‘Unread’ notifications have the highest priority
  • A strong focus on on action and date
  • Minimal variation on notification type

With this information I could move to establishing a stronger hypothesis and validate the opportunity with my data analyst.

Hypothesis

Improving visibility and organization of notification center will ensure users find value and actionability within it. This will lead to higher retention by providing stronger direction and prioritization of efforts for users.

Opportunity

  • Increase team weekly active users to 43%, increase of 4%
  • There are 2.2M+ active users weekly, both on paid and free plans.
  • About 14% of these users are active in team workspaces, if we can increase this by ~4% we would see an additional 13,000 active users in team workspaces a week or 670,000 a year.

Defining our goals

Once we had our hypothesis and opportunity mapped out I aligned our business goals to user goals.

User Goals

  • I am aware of important changes within my Postman system
  • I can quickly take action on high priority events

Business Goals

  • Increase retention and team conversion through notification visibility

Before : Where we were

Circling back on where we were, the notification center at Postman was:

  • Ordered by time
  • Quality varied pending the template used
  • No way to view and/or search
  • No prioritization provided
  • Hard to discover

Defining a Notification

To better ensure the new notification center pattern extends beyond this initiative as well as solves user needs I spent some time defining what they are and how they should work.

Notifications serve as methods to inform users of system status’. Often as a means to create action from them, whether that action is immediate or informative.

Types of notifications:

  • High priority – Immediate action needed
  • Medium priority – Act on at some point
  • Low priority – Muted, no action needed

Notification method:

  • Global : visible anywhere in-app
  • Contextual : within context of a space
  • External : outside app methods (ex: email)

Low-Mid Fidelity : Invite explorations

I explored ~5 variant, each focused on how I could better align in-app notifications to a users expectations and ensure they are able to action on high priority alerts. As you look over this you’ll see more refined approaches as I aimed to match the old world with the new. This work would have a deep focus on these factors:

  • Build trust through system visibility
  • Bring awareness not noise
  • Expose users to a community over silo
  • Provide actionable next steps pending notif priority

High Fidelity – The A/B Test

Below is the solutions I created to help users stay up to date on recent changes and take action on high priority requests. To ensure success I:

  • Validated designs through crits and reviews
  • Collaborated with the engineering team to ensure quality of UX
  • Established animation needs with engineers
  • Tracked success through growth experimentation timelines

Flow – Receiving a notification and engaging with Notification center

This flow shows what a user sees when a new high priority notification arrives. A method of bringing users attention was integrating a pulsing animation that helps, in a non-loud way, alert them. After a few pulses it turns back to a standard orange circle and waits for them to engage. Along with this I had reorganized notifications into different categories to ensure easier consumption.

Below you will see a figma prototype of this work, if you need to reset it press ‘R’.

Experiment Results (Mar 10, 2023)

The experiment ran through Feb 2023. The below results showcase the impact it had and why it would it would be pushed to 100% of our users (aka GA).

13%

Increase in-app notification usage

15%

Increase in invite acceptances

Additional data found from the study was:

  • Pulse indicator performed slightly better then the no-pulse
  • Impact grew steadily over time, not huge changes in first week

This showed our team that the changes to the ‘notification center’ had a positive impact and could lean to longer term life improvements for our users.

Road to GA

With the success of the experiment, the next step was optimizing the overall design for GA:

  • I worked closely with my product partners to share and promote findings
  • Made some final decisions about what we were going to ship
    • Some tweaks to UI to better match Postman branding
    • Working with partners to ensure new notifications follow this pattern
    • Final evaluation of notifications within system
  • Kicked off planning the next set of experiments to enhance impact further

Original Invite Modal

New domain invite modal

Outcomes

GA’d in May 15, 2023

Has seen steady growth in usage

Notification bell clicks per month grew by 331%

Established new notification pattern throughout Postman

Larger re-evaluations of notification and user campaigns