Dashboard
Leaders start with the customer and work backwards. They work vigorously to earn and keep customer trust. Although leaders pay attention to competitors, they obsess over customers.
dashboard (Feed 2 birds with one scone)
Situation:
- Service Engineers often have a lot of cognitive load and need to merge information quickly when dealing with incidents and monitoring. There are many diverse systems that have to look at - Pagerduty, Grafana, Kibana, Slack, Github, deploy systems, etc.
Task:
- Build a system that solves these problems. Focus on service engineer persona. What do they care about?
Action:
- Design
- Spec Writing
- Timeline view
- FE - cross links scattered everywhere, think about error handling in the case something (external or internal) is not working
- BE - interface with APIs, manifest stuff, other upgrades for systems
- github cdn?
- documentation, guides for platform etc
Result:
- "Makes me want to own more services" - an engineer
- used by almost half of population throughout a month
- value from the exploratory aspect as well as the incident aspect
- cannot really measure cognitive load, but can measure user happiness
- Think about the different perspectives of incident triage vs service exploration
Reflection:
- What could be improved? Reaching out to maintainers of OS libraries (to uncover issues earlier)
- Estimation (Milestone doc quickly went out of date)
- User research / TPM / EM resources can be useful
5 Whys:
- uwu
Q&A:
- uwu