Incident Automation Problem Validation
Problem Statement
Incidents contain too much information and not enough relevant context
Reach
This work with reach users who are responsible for the availability and reliability of the applications they work on. The primary persona is Devon the DevOps Engineer and the secondary persona is Sasha the Software Developer.
There does not exist exact data of the percentages of our user base who fill the personas listed above. We can say with confidence majority of the users on our platform are Developers and a much smaller number have the role of DevOps Engineer or Operations Engineer. This feature will mostly serve those who work in operations although it will also be used by developers at more progressive companies who are involved in running the code they write and are on-call for it.
Based on the assumptions listed above, it has been estimated that work for Incident Automation will have a small reach - 1.5 (~5% - ~25%).
Impact
Solving this problem streamlines the initial investigation phase of the Triage process. It helps the on-call responder to quickly diagnose the problem and take first action. This has the following positive effects:
- Decreases time spent
- Decreases alert fatigue
- Boosts moral
- Decreases $$ lost in an outage
For personas experiencing these problems, the impact is High (2.0). For GitLab's business, the impact is Low (0.5) - it is lower because GitLab's operations platform is still at early maturity. This work is critical to building out a robust operations platform, but in and of itself, it will not have immediate ROI.
Confidence
TBD
Effort
TBD
References
Opportunity Canvas
UX Research Issue
Customer Interview Guide
Incident Automation Work by Internal Infra team