Back

 
 

Listening to Users to Revamp an Alerting Workflow

Problem

Our users avoided using Tanzu Observability’s out of the box alerts because the experience was disjointed and didn’t offer sufficient customization options.

Goal

Increase the installation rate of our out-of-the-box alerts to help users derive more value from the product from day one.

Team

Product owner, Software engineers (3), Lead UX Designer (Me)


Background

Company

With over 80+ products, VMware (now Broadcom) is a leading provider of multi-cloud services for applications. VMware helps provide the infrastructure for the applications that we as UX designers work on every day.

Site Reliability Engineers

Behind the scenes, Site Reliability Engineers (SREs) are the unsung heroes working tirelessly to ensure websites remain up, responsive, and efficient. This is no small task, which is why SREs rely on powerful tools to monitor and maintain system health. Enter observability—the key to giving SREs the visibility they need to troubleshoot issues and optimize performance to keep websites running smoothly.

Observability

Observability tools help developers, SREs, and platform operators gain insights into the systems that power their applications. During my time at VMware, I spent my time working on Tanzu Observability, an observability tool that brings data from multiple sources into one platform. It enables users to create custom dashboards for real-time visibility, set up alerts to notify site administrators of outages, and drill down into traces and logs to identify the root cause of incidents.

 

Discovery

Alerting is the function of a monitoring system to detect and notify operators of critical events that indicate a significant change in system status. Tanzu Observability has out of the box alerts a user can “install” to hit the ground running from day one. User adoption of our out-of-the-box alerts was suffering from low adoption, which led our product team to focus on addressing the gap.

Discovery Calls

With support from customer success, I organized discovery calls with five of our top customers to better understand their experiences. Here’s what we learned:

Our biggest problem was the inability to template alerts
The alerting workflow is disjointed and time consuming
Using out of the box alerts was a pain because you can install them, but you can’t edit them.

Problem Statement

As a user of Tanzu Observability, I avoid using the out of the box alerts because the experience is disjointed and I don’t have the ability to customize the alerts to work for my business needs.


Sharing Insights Across Teams

To make sure the insights from our discovery calls were shared across the team, I organized a research readout session with the entire product and UX organization. It was a great way to spread the knowledge and get everyone on the same page.

My work on out-of-the-box alerts touched multiple product teams, so I pulled in the right designers and product managers from the get-go. Their unique insights became the project’s guiding light, and I set up regular check-ins to keep the feedback loop flowing and ensure we stayed on track.


Ideation

After defining our problem statement, it was time to start crafting a solution.

Led a cross functional workshop

I led a cross functional workshop including product management, UX and engineering to ideate around solutions to the pain points we addressed in our discovery.

Our design direction

Create an intuitive, scalable, and customizable alert template flow that would minimize time to value for users.

User flows

I crafted user flows that would support the ability to create alert templates. These deliverables were used to communicate, socialize and validate our broader vision with stakeholders.

Wireframes

Once we iterated on our user flows, I continued by creating more detailed wireframes. Focusing on what we could deliver in the short term and socializing further.


Prototype

After gathering feedback throughout the process, I designed a high-fidelity prototype for user testing.


Test

The moment of truth: Did our designs really improve the user experience?

Research Questions

  • Do users understand the alert template concept and is it discoverable? 

  • Do users know how to create an alert from a template? 

  • How important are bulk actions while creating alerts?  

I created a Miro board with the questions on the left and customer responses on the right. This setup allowed us to take quick notes during the calls and easily compare how the different customers answered each question.

Learnings

Users wanted to be able to edit the entire alert query instead of just the thresholds.

There was a lack of discoverability — users didn’t understand how to get to the alert templating flow.

Users wanted the ability to create their own alert templates instead of just out of the box options.

5/5 Users found adding alerts from templates in bulk very useful


Next Steps and Learnings

Our testing proved we were on the right track, but we needed to make a few adjustments to the prototype to address some of the pain points that came up during our calls.

  • Add the ability to edit the query when creating alert from template 

  • Create a custom template workflow in addition to out of the box templates

  • Improve discoverability by adding alert templates to the main navigation

After refining the design for development, the project was deprioritized. Despite our obvious disappointment, we documented everything for future use.

A year later, the product team asked us to replicate the alerting functionality in a new product. Revisiting our research and designs gave us a strong case for rebuilding the experience the right way, and the product strategy shifted in our favor.

The lesson is clear: research is never wasted. Even if results aren’t immediate, the insights can drive future success, and properly documenting and sharing findings ensures they’re never lost.

 
Previous
Previous

Refining Design Through User Feedback