
By Shane Tyrrell - Head of Advisory and Capability Enablement, Prosci ANZ
A retail organisation has just cut over to a new ERP system. The technical migration was flawless with the data reconciled to the cent and user acceptance testing signed off two weeks early. The project manager stood in front of the steering committee the day before go-live and reported a textbook project execution - crack open the champagne!!
By the end of the first week post go-live half the finance team was running month end in Excel again alongside the new system because they did not trust the numbers it produced. 3 people had built private macros to recreate reports the old system used to generate automatically. 1 team lead told her staff to “use the new system for the audit trail” and keep doing the real work the old way.
The system worked exactly as built. The cutover checklist had fifty line items, not one of them asked whether the people using the system were ready to use it.
If you have a go-live on your project plan this month or quarter, this story should make you uneasy. Because the research data says it sits closer to average user experience than being the exception.
The checklist that goes missing
Most go-live checklists are built around the technology. Data migration validated, integrations tested, security sign-off complete, rollback plan documented. All necessary steps on the journey yet none of them ask whether the people on the other side of the system are ready to use it.
Prosci’s 2025 Unlocking ERP Implementations research study found that human factors matter 6x more than technical factors in determining whether an ERP delivers its promised value. Not slightly more important, 6x.
6X
how much more human factors matter than technical factors in ERP value realisation and success, Prosci’s 2025 research
Most budgets still run the opposite way. The average ERP project allocates 92% of its budget to technical work and 8% to change management according to Prosci’s Best Practices in Change Management research.
92% / 8%
the average technical versus change management budget split on ERP projects
Gartner’s research points to where that imbalance leads. By 2027 more than 7 in 10 recently implemented ERP initiatives are projected to fail to fully meet their original business case. Not because the software stopped working but because the organisation around it was never made ready to use it.
Testing the system isn’t testing readiness
UAT proves the software does what it is supposed to do. It proves nothing about whether people want to use it, understand why it matters or can perform under the pressure of a live environment instead of a sandbox. Insufficient user testing shows up repeatedly in Prosci’s research as a leading cause of ERP failure not because teams skip testing but because testing measures whether the system passes, not whether the people are ready.
That gap is exactly where the finance team above tripped up. Every test case passed but nobody tested whether the team trusted the output enough to delete their spreadsheets.
The check itself
Run this against every impacted group two weeks before go-live, not on launch day. Five short questions mapped to ADKAR, that will tell you more about readiness than any test script.
- Awareness: can someone in this group explain in their own words why this change is happening and what happens if it doesn’t?
- Desire: are they choosing to adopt this or being told to?
- Knowledge: have they used the system on their own data, in their own role and not a generic demo?
- Ability: have they completed one full task end to end under realistic time pressures?
- Reinforcement: is there a named person checking adoption in week one or does the plan stop at go-live?
A “no” to any of these questions is not a note for the comms team to action later. It is a go or no-go signal carrying the same weight you would give a failed UAT test case.
Three things to do before your next go-live
First, run the check on a representative sample of users not your super users. Super users were trained first and involved longest which makes them the least useful barometer of where the rest of the organisation sits. Pull from the middle of the curve, the manager on leave during training and the team running the most workarounds in the current system.
Second, treat amber and red answers the way you treat a failed test case. Escalate them, assign an owner and set a date to close them out. Most readiness gaps get raised in a steering committee slide and forgotten by the next meeting. Run them through a defect log instead and they get fixed.
Third, name a reinforcement owner before go-live, not after. Someone whose job in week one is to watch for the macros, the workarounds and the team lead quietly telling people to do it the old way. That person needs to exist before launch day and not get appointed once adoption has already slipped.
What this means for you as a PM
Your go-live checklist almost certainly has fifty boxes for the technology. Data migrated. Integrations tested. Rollback plan documented. Security sign-off complete etc.
This is not a request for you to become a change practitioner. It is an ask to add five questions to a checklist you already run two weeks before a date you have already set. Go-live is not the finish line, It is the point where you find out whether you built readiness alongside everything else.



