You need to sign in or sign up before continuing.
2021-09-16 FRT SLA / NRT SLA Working group
Date Formed: 2021/09/16
Date Ended:
Proposed Exit Criteria:
3 weeks of above 93% SLA on Self-Managed and SaaS (as Saas account L&R are not seeing a problem)
Members
Member | Role |
---|---|
@lbot | Facilitator |
@Ikosenko | Member |
@cynthia | Member |
@greg | Member |
Which metrics are below target?
FRT SLA / NRT SLA
Related to https://gitlab.com/gitlab-com/support/support-team-meta/-/issues/3746
Following https://about.gitlab.com/handbook/support/managers/#forming-a-group
@rspainhower rightfully suggested a review group
-- What happened in AMER? -- Is this temp? -- Is there a change that needs to be made
Ilia has some analysis here: https://gitlab.com/gitlab-com/support/support-team-meta/-/issues/3746#note_679511651
Screenshots
Why are these metrics below target?
-
HypothesisHigh Confidence HypothesisIn Progress Hypothesis - @cynthia
- Regional Split is causing ticket overflow into handover hours which leads to breaches - Likely -
HypothesisLow Confidence Hypothesis - @cynthia
- Increased Escalations during this period has lead to FRT or NRT SLA decrease - Correlated -
HypothesisHigh Confidence HypothesisIn Progress AMER-W continues to be a hotspot for SLAs due to time+space - @greg -
HypothesisHigh Confidence HypothesisLow Confidence HypothesisIn Progress Hypothesis 2 - @Ikosenko - Increased Non-Ticket activities (PTO/Training/360s) during July and August may have contributed -
HypothesisHigh Confidence HypothesisIn Progress Prioritizing FRT over NRT led to a decrease in NRT - @greg -
HypothesisLow Confidence Hypothesis - @cynthia
- Potentially a "Monday" problem - Debunked -
HypothesisLow Confidence @lbot
- HPAR tickets are having lower SLA performance with the new workflow - Debunked -
HypothesisLow Confidence Hypothesis @lbot
- Why are most tickets breaching on Wednesday -- Tuesday is highest ticket creation day which leads to high breaches Wednesday - Debunked
Need Owners
- [] Hypothesis
@lbot
- Was there a change in ticket composition over this time (14.0 problems?) -
HypothesisHigh Confidence HypothesisLow Confidence HypothesisIn Progress Hypothesis 1 - @assigned-member
- Increased page-events (CMOC/Emergencies) lead to engineers swarming those issues has lead to FRT or NRT SLA decrease -
HypothesisHigh Confidence HypothesisLow Confidence HypothesisIn Progress Hypothesis 2 - @assigned-member
- Hiding Assigned tickets has reduced the average number of tickets resolved per engineer. (PTO may throw this off)
Untestable
-
HypothesisHigh Confidence HypothesisLow Confidence HypothesisIn Progress Hypothesis 2 - @assigned-member
- May not be testable: C O V I D C H A L L E N G E S (Hard to measure but need to think of it) -
HypothesisHigh Confidence HypothesisLow Confidence HypothesisIn Progress Hypothesis 2 - @assigned-member
- May not be testable: Removal of SLA Hawk in APAC means less AMER tickets get saved. -
HypothesisHigh Confidence HypothesisLow Confidence HypothesisIn Progress Hypothesis 2 - @assigned-member
- May be superseded: Hiding Assigned tickets has increased NRT, RWT and TTR, and decreased NRT SLA performance
What actions are being taken to address these?
-
Action 1 - @assigned-member
- Result Summary
Artifacts created as a result of this working group
Screenshot of metrics at or above target for 2 weeks
What process changes resulted from this workgroup?
What new hypothesis and analysis resulted from this workgroup?
What other handbook updates resulted from this workgroup?
What other information would be interesting to note (but not capture in an MR)
Meeting Records (if any exist)
Edited by Lee Matos