Issue page
Issue resolving flow for New Relic AIerts & AI
OVERVIEW
Incident management aims to identify and correct problems while maintaining normal service and minimizing the impact on the business.
An important part of developing software is addressing incidents quickly. Engineers at every level couldn't underappreciate efficiency in building, testing, and deploying software. Investigating issues is just the tip of the iceberg. The full workflow will consist of setting monitoring tools and configuring problem detection and notifications. Depending on the product, size of the organization, and management style work might be organized differently so the monitoring tool might consist of the maximum possible flexibility and compatibility with other tools.
In November 2021 New Relic presented a new tool that is solving this problem and I took the lead on its design.
Incident management aims to identify and correct problems while maintaining normal service and minimizing the impact on the business.
An important part of developing software is addressing incidents quickly. Engineers at every level couldn't underappreciate efficiency in building, testing, and deploying software. Investigating issues is just the tip of the iceberg. The full workflow will consist of setting monitoring tools and configuring problem detection and notifications. Depending on the product, size of the organization, and management style work might be organized differently so the monitoring tool might consist of the maximum possible flexibility and compatibility with other tools.
In November 2021 New Relic presented a new tool that is solving this problem and I took the lead on its design.
MY ROLES
+ Product Designer
+ Researcher
PROJECT GOAL
+ Consolidated flow troubleshooting flow for the whole issue lifecycle
+ New feature integrated
+ Product Designer
+ Researcher
PROJECT GOAL
+ Consolidated flow troubleshooting flow for the whole issue lifecycle
+ New feature integrated
Issue lifecycle. User flow
Challenge
Reduce cognitive load and increase problem solving efficiency
Incident management is the least favorite job of DevOps by default. It easily became overwhelming and it's a never good time to be notified about the problem. So reducing the noise and proper configuration of rules and setting notifications is critical. So the main challenge was to structure multiple features and complex configurations to reduce cognitive load on every stage of the user flow.
USERS
The users of the Alerts and AI could have different roles and titles, they could interact with a full workflow or just with some products or features. Some of them might set correlation rules or configure notification channels, but others might be just investigating problems during their on-call shift.
The users of the Alerts and AI could have different roles and titles, they could interact with a full workflow or just with some products or features. Some of them might set correlation rules or configure notification channels, but others might be just investigating problems during their on-call shift.
RESEARCHES CONDUCTED
+ Interviews
+ Card sorting
+ A/B testing
+ Recordings analysis
+ Usability testing
+ Interviews
+ Card sorting
+ A/B testing
+ Recordings analysis
+ Usability testing
DevOps
Planning responses to potential incidents by identifying weaknesses in systems. Setting up monitoring tools, alert systems, and runbooks that help team members deal with a problem.
On-call engineer
Getting notifications during the shift needs to react and analyze the issue. Then if possible – to detect and solve a problem.
Observability engineer
Application monitoring to detect problems and anomalies.
Finding the root cause of the issue and getting insights for better performance.
Finding the root cause of the issue and getting insights for better performance.
Tech lead/team manager
Managing team observability process, on-call shift, and analyzing the efficiency of solving problems.
Issues view
The landing page of the product gives an overview of the problems of the services, applications and DBs, connected to New Relic.
The goal of this feed is to organize issues. Users can find a short summary of all the problems, and compare or analyze dynamics. According to user data, the most important parameters were organized in the table. Sorting and filtering include selectors that could help different types of users with different goals and roles to organize issues according to their needs. For example, team managers are concentrating on improving the efficiency of the team monitoring processes, and on-call engineer might use flexible filtering for finding issues by one or more multiple parameters.
Issue feed view as a landing page of the product
Issue page
Provides insights to help users first understand the problem, and then minimize the time to resolve it.
The issue page is a gateway to the important data that would help to troubleshoot. It contains all the possible information New Relic can provide according to the actual issue. It contains information about the timing, notifications sent, full information about all the related incidents, entities (services, applications, DBs, etc), error logs, dashboards, issue timeline, and other features. The main challenge was to organize all the types of information in a way that would be intuitive for most of the users and not overwhelming. The multiple research initiatives helped to understand the journey map of the different clients better and adjust information architecture to users' needs.
Issue page variation
Issue feed and Issue page at iOs NR1 application
Issue title
Incidents list
Outcome
Opportunities became challenges
Issue-resolving workflow project was going together with product experience consolidation and big migration to a new pipeline. Issues feed and page are the central pages of the product and from the first day we opened it to a small amount of users we were getting a lot of feedback. The new design brought a lot of attention and was getting throw changes consistently. The page usage increased dramatically comparing the previous version. Users got much more data that might be helpful for troubleshooting, but on the other hand, IA became more complex and harder for scanning. So the next step was obviously to work on customizing the Issue view automatically according to the user's work patterns.
Product Designeress
Residential Proxy Service
Custom Medieval Costumes, Armor and Armory Store
Sell & Buy map-based application
