I've walked into a lot of companies where the ERP migration was supposed to be finished two years ago. The old system is still running. The new system is partially implemented. And the accounting team is doing double data entry into both systems indefinitely.
Nobody planned for this. But here we are.
The Pattern
The story usually goes something like this: the company outgrew its legacy system. Leadership decided to modernize. A vendor made compelling promises. The project kicked off with enthusiasm.
Then reality set in.
The data migration was messier than expected. The customizations took longer. The training didn't stick. Go-live kept getting pushed back. Eventually, the project lost momentum, key people moved on, and the company found itself in a state of permanent transition — which is to say, a state of permanent dysfunction with two systems instead of one.
Why This Happens
In my experience, ERP migrations stall for predictable reasons. What started as a system replacement quietly becomes a complete business process redesign — both valid, but very different projects with very different timelines. The data has twenty years of decisions embedded in it, some of them wrong even at the time. Implementation partners want to sell hours, internal IT wants technical elegance, finance wants accurate numbers, and operations wants everything to keep working like it always has. Someone needs to arbitrate those tensions, and that someone is often missing.
And then there's the definitional problem: nobody agreed on what "done" means. Is the migration done when the software is installed? When the data is migrated? When everyone is trained? When the old system is decommissioned? If you can't answer this question, you can't finish the project.
What To Do Now
If you're stuck in a stalled migration, here's where I'd start.
Acknowledge the situation. You're not finishing the original project. That ship has sailed. What you're doing now is triage: figuring out what's working, what's broken, and what's the minimum viable path to solid ground.
Get honest about the business impact. Double data entry isn't just inefficient — it's a source of errors, a bottleneck in your month-end close, and a morale killer for your team. Quantify it. That gives you leverage to get the resources you need.
Define a new finish line. Maybe it's not the full implementation you originally envisioned. Maybe it's getting 80% of the value with 20% of the remaining effort. What does "good enough" look like? Answer that and you have a project again.
Get outside perspective. Not another vendor. Not another consultant with something to sell. Someone who can look at both the technical and business reality with fresh eyes and tell you the truth about where you stand.
The Path Forward
Stalled ERP migrations aren't hopeless. I've inherited several and gotten them to a workable state. The key is accepting that you're not starting over — you're finishing what was started, with clear eyes about what's actually achievable and a plan that doesn't require anyone to be a hero.