Hi everyone,
In 2025 I delivered four Jira Data Center to Cloud migrations, and afterwards I noticed that users had not reacted to them in the same way. At some clients people carried on as if nothing had changed, while at others activity stayed lower for weeks. I turned that into my master's research, and I'd like to share what came out of it and hear how others handle this part of a migration.
How we measured it
Rather than surveys or interviews, I used the Jira changelog, because it shows what people actually did rather than what they remember. I counted people who made at least one change on a workday, not people who logged in, after removing bots, app accounts and automation rules. I also compared each person against their own activity before the migration, so the result didn't rest on the site total alone.
Before calling a drop a problem, I ran the same calculation on ordinary days before the migration. Activity swings a lot on normal days, especially on small instances, and I found no published benchmark for what normal looks like.
What we saw
One client looked recovered within two weeks, with the number of active people back near normal. But only 37% of regular users were still doing at least four fifths of their usual work, against 59% at another client measured the same way. Workflow transitions, field edits and worklogs were all affected, so it was not just a few advanced features. How deep the dip goes and how fast it recovers also behaved as two separate things, since one client fell further than another and still recovered sooner.
The only migration with no measurable dip was also the only one where users could actually reach a person at the moment of the switch, through an organization-wide war room with champions from the business on first line. Another client had a dedicated hypercare project in JSM, and not a single end user opened a ticket in it. The technical cutover was solid at all four, so that wasn't what made the difference.
It's four cases, so these are patterns, not statistical proof. I wrote up the full story here: https://lnkd.in/p/emv8Tv6n
Questions for you:
- Do you measure adoption after a Cloud migration, and if so, with what?
- How do you make sure hypercare actually reaches end users and not only IT staff?
- Does anyone keep some form of adoption governance going after the project closes, given how often Cloud changes?
Curious to hear your experience.
Narek
