Measurement, Continual Improvement and Organizational Change Management
What to measure
Good change metrics answer whether the practice is enabling successful change, not just how busy it is. Useful measures include: proportion of changes assessed with a documented risk evaluation; change success rate (changes closed without causing an incident or rollback); count and ratio of emergency and unplanned changes; lead time and cycle time from registration to implementation; number of changes rejected or deferred and why; backlog and throughput of changes; and change-related incidents as a share of total incidents. Vanity metrics - number of changes processed, number of CAB meetings held - say nothing about value.
Continual improvement loop
Measurements feed the improvement register: are standard change models still valid, is categorization calibrated against actual failures, are approvals happening at the right level, is the schedule causing avoidable conflicts? Failed or rolled-back changes are the richest source of improvement, and standard change models should be reviewed against real failure data.
Organizational change management
The people side matters as much as the process. Organizational change management (a general management practice) addresses stakeholder engagement, communication, training, sponsorship and resistance. In change enablement terms: people are more likely to adopt a change if they were engaged early, understand the reason for it, and have been prepared and trained. A technically perfect change that nobody adopts is a failed change. Questions often present a scenario where the change worked technically but users reverted to old practice - the answer lies in OCM, not in more testing.
Traps
Don't confuse change management (the OCM discipline, people side) with change enablement (the ITIL practice, lifecycle and risk side); older terminology causes deliberate confusion here. And don't propose adding approval layers as the fix for a poor success rate - root cause is usually assessment quality, categorization or scheduling.
Rule of thumb: for every metric ask "what decision would this change?" If none, stop reporting it.