Two releases a year – every year
ServiceNow ships two major versions, known as family releases, every year. Up to Zurich they were named after cities in alphabetical order. With the alphabet used up, ServiceNow switched to countries in 2026: first Australia, then Brazil.
That pace means upgrades are not a one-off project but a recurring process. The organisations with the least stress are the ones that have made upgrading a routine.
1. Read the release notes – focusing on what you actually use
Release notes are extensive. Start with the applications and modules you have activated, and look especially for:
- features being removed or replaced (deprecations)
- changed default behaviour in forms, workflows and portals
- new features that could replace your own customisations
If you skip a release, combined release notes show the changes between two versions.
2. Review your customisations
What usually causes trouble is customisation of baseline functionality. The more you have changed ServiceNow's own files, the more skipped records you get during the upgrade – records where the platform does not overwrite your version.
Use the upgrade as a chance to ask: is this customisation still needed, or does the baseline now do the same thing?
3. Test on a fresh copy of production
Clone production to a test instance and upgrade that first. That way you test against real data and real configuration.
- Automated Test Framework (ATF) lets you retest critical flows in minutes instead of days. If you have no ATF tests, the next upgrade is a good moment to start with your ten most important flows.
- Instance Scan helps you find known issues and deviations from best practice before you upgrade.
- Don't forget integrations. Test them against the actual connected systems, not just that the calls go through.
4. Handle skipped records methodically
After upgrading the test instance, go through the list of skipped records and decide for each one whether to keep your version, take ServiceNow's new version or merge them. Document the decisions – the next upgrade will go faster.
5. Plan the production upgrade
- Pick a low-traffic window and communicate it well in advance.
- Have a clear plan for verification right after the upgrade.
- Decide beforehand who makes the call if something does not work as intended.
Our experience
Upgrades that drag on almost always have the same cause: lots of customisation of baseline functionality and too little automated testing. It pays to invest in both between upgrades, not during them. An instance close to baseline with good ATF coverage can often be upgraded in a few weeks instead of a few months.