Solution validation for Custom Integrations
What's this issue all about? (Background and context)
As part of gitlab#217766 (closed), we are designing a way for people to build custom integrations. Custom integrations would allow users to map fields in their alerting tools to GitLab alerting fields, essentially allowing them to better display their alerts within GitLab.
Does what we're proposing make sense? How can we continue to improve this offering, in future?
What are the overarching goals for the research?
- To see if the solution we are proposing is usable.
- To find opportunities for improving the experience in both the short and longer term.
- To get a better sense of how people might want to use this functionality, when it's in place. For instance, will they want to share the integrations?
What hypotheses and/or assumptions do you have?
- That users are going to want to build custom integrations
- That having a simple, UI-based mapping option is preferable to users (as opposed, say, to mapping fields in code)
What research questions are you trying to answer?
- What experience people have setting up alerting integrations?
- What information they need to set up an integration properly?
- How they will use an integration, once it's established.
What persona, persona segment, or customer type experiences the problem most acutely?
DevOps Engineers, Developers and System Administrators
What business decisions will be made based on this information?
If we should build the custom integration builder as currently defined.
What, if any, relevant prior research already exists?
In previous studies, we have seen that our original way of setting up integrations was confusing. Hopefully this is an improvement on that original effort.
Who will be leading the research?
Amelia
What timescales do you have in mind for the research?
Needs to be completed in 13.3
Relevant links (script, prototype, notes, etc.)
Edited by Amelia Bauerly