Post-incident review solution validation
What did we learn?
Results |
---|
Process can be improved by ensuring the review template is easier to add and update, and so it's easier to see the "create review" button, but the general idea for creating a post-incident review is sound. Future improvement ideas include embedding an image of the timeline, automatically adding participants and incident time/date to the retro issue, and automatically emailing people involved in the incident when the review issue is created. |
Dovetail project |
What's this issue all about? (Background and context)
In https://gitlab.com/gitlab-org/gitlab/-/issues/360439, we identified an approach for a first iteration of post-incident reviews. We want to verify that our approach is valid and usable for people using incidents.
What hypotheses and/or assumptions do you have?
That what we're proposing is a reasonable way to introduce post-incident reviews to GitLab.
What questions are you trying to answer?
General questions
- Does the general workflow make sense?
- Does this approach add value to the product?
- Are the ways we're surfacing the post-incident reviews to people useful (or just annoying)?
- Are there things we could do that could make this better (ex ways to automate the population of the review issue)?
Specific questions
- Does it make sense to add a link to the timeline to the post-incident review issue?
- Does it make sense to mention or assign to participants/assignees automatically?
- How would people expect to interact with the action items list? What do they think they could do from the action items list (create a new issue, to-do item, task, or something else)?
What research methodology do you intend to use?
Unmoderated testing sessions with at least 5 external participants.
What persona, persona segment, or customer type experiences the problem most acutely?
Ingrid, Infrastructure Operator
What business decisions will be made based on this information?
Whether to proceed with the MVC proposal for post-incident reviews.
What, if any, relevant prior research already exists?
Problem validation conducted as part of https://gitlab.com/gitlab-org/ux-research/-/issues/1925+
Who will be leading the research?
Amelia
What timescales do you have in mind for the research?
To be completed in 15.7
Relevant links (problem validation issue, design issue, script, prototype, notes, etc.)
- Design issue --> https://gitlab.com/gitlab-org/gitlab/-/issues/360439+
- Problem validation issue --> https://gitlab.com/gitlab-org/ux-research/-/issues/1925+
TODO Checklist
-
Decide on approach -
Create Dovetail project -
Create survey on usertesting.com -
Launch and run with 5 participants -
Analyze results -
Add insights to Dovetail
Edited by Amelia Bauerly