How to Choose a Windchill Upgrade Partner

A Windchill upgrade is mostly invisible when it goes well and very visible when it doesn’t. The partner you pick decides which one you get. Here are the questions we’d ask, and what good answers sound like.

1. How many rehearsals, and what counts as “done”?

Ask how many full rehearsals they plan and what the stopping rule is. “Two rehearsals” is a budget. “We rehearse until two consecutive runs finish clean” is a standard.

The industry norm is one or two rehearsals because each one takes days of skilled manual work. On a recent Windchill 12 to 13.1 upgrade we automated most of the steps, ran eight rehearsals, and stopped after two clean runs in a row. Every rehearsal failed somewhere new, which is exactly why it was worth running the next one.

2. What’s the downtime window, and how was it proven?

You should get a specific production window (usually a weekend) and evidence from rehearsals that the cutover fits inside it. If the answer is an estimate with no rehearsal behind it, keep asking.

3. How will they handle our customizations?

A largely out-of-the-box Windchill is one job. A heavily customized system that has been running for ten years is a different one. A good partner will want an inventory of customizations early and tell you, for each one, whether it survives the version change, needs rework, or should be retired.

4. What’s the rollback plan?

Ask what happens at hour 30 of a 48-hour window if something fails. There should be a written decision point and a tested way back.

5. How will they verify the result?

“It came up” isn’t verification. Ask what they’ll check after cutover: data counts, workflows, integrations to ERP, CAD connections, user access. Ask to see the checklist.

6. Who’s working the weekend?

Find out who will be on the cutover, how senior they are, and whether they ran the rehearsals. The people who rehearsed it should be the people running it.

7. What do we keep afterwards?

A runbook, an error catalogue, and a list of configuration drift found and fixed. On our 12 to 13.1 upgrade, repeated rehearsals surfaced old settings and misconfigurations that had nothing to do with the upgrade itself; they were fixed before production instead of being discovered during it. That record also makes the next upgrade cheaper.

What good looked like on one project

The AI-verified Windchill upgrade is our worked example: a two-hop upgrade from Windchill 12 to 13.1 on a clustered production system. All 40 of 40 migration phases ran with zero failures and zero production outage. The tradeoff is upfront work: the automation, rules and error catalogue had to be built before the first rehearsal, which a conventional plan doesn’t need. For a similar environment, we’d now expect three rehearsals or fewer.

Frequently Asked Questions

How much downtime should a Windchill upgrade need?
It depends on the size and customization of the system, but production downtime should be planned to fit a defined window, usually a weekend, and proven in rehearsal first. On a recent 12 to 13.1 upgrade we ran, all 40 migration phases completed with zero production outage.

How many rehearsals should a Windchill upgrade have?
Enough that the last runs finish with no surprises. The industry norm is one or two because each one takes days of manual work. When rehearsals are automated you can run more; on one upgrade we ran eight, stopping after two consecutive clean runs.

Can we skip a version, for example go straight from Windchill 12 to 13?
Some paths need an intermediate step. A two-hop upgrade is common and should be planned and rehearsed end to end, not as two separate projects.

What should a partner deliver besides the upgraded system?
A tested runbook, a rollback plan, a list of customizations and how each was handled, and a record of configuration problems found and fixed along the way.

Planning a Windchill upgrade?

Element Consulting has upgraded, migrated and rescued Windchill since 2003. We’ll tell you what we’d rehearse first.
Talk to a Windchill Expert