Definition (Hover Text)

Mean time to Restore (MTTR) is a metric that measures the average time taken to recover from production incidents. It is calculated based on the production incident tickets raised during a certain period of time.

Source Tools

Jira

Graph type

Line

• Y- Axis - Hours

• X- Axis - Weeks

• Positive trend indication - Decrease

Filters

N/A

Hover Format on KPI

<XX> Hours

Fields on Overlay

• Project Name

• Date

• Story ID

• Issue type

• Issue Description

• Creation Date

• Completion Date

• Time to Recover

Business Logic

Calculation Formula

Diff between Closure Date and Creation date of all issues in a week gives the average MTTR for a week.

The average of 8 weeks is shown in the centre column

Trend

Increase in MTTR is a -ve trend

Maturity Levels

M1: >1 Week 

M2: 1 day - 1 week,

M3: 12-24 Hours,

M4: 1-12 Hours,

M5: < 1 Hour

Instance level thresholds

N/A

Configurations

Processor Fields

None

KPI Specific fields

Production incidents identification - Custom Field or Labels

Issue type to identify Production Incidents

DOD Status

How to Validate KPI

Suggested ways of working

Team should either have a separate issue type to identify production incidents or they could use a custom field or add a label to identify them

As soon as a production incident happens, an issue should be created in Jira. Once the issue is resolved, the corresponding issue should be closed. The difference between creation and closure of the issue is MTTR

Sample JQLs

Benefits of KPI

How does the KPI help

The KPI helps to gauge the agility with which the a team restore an application to normalcy