Your ERP went live a year and a half ago. On paper the project is done: the system runs, the team got trained, the consultants sent their final invoice. But month-end close still happens in Excel. The inventory counts on the floor don’t match the counts on the screen. And when you call the partner who built it, three days go by before anyone calls back.
None of this means your team failed. In most cases, it doesn’t mean you bought the wrong software either. But before you spend another dollar, you need the answer to one question: is the system broken, or is it unfinished? Everything that comes next depends on which one it is, and most of the people offering to help you have a financial interest in getting the answer wrong.
Why this happens to good companies
ERP projects are scoped, budgeted, and managed to reach go-live. Go-live gets treated as the finish line, when it’s really the point where the hard part starts. When timelines slip and budgets tighten, and they almost always do, the pieces that get cut are the ones that come after the system turns on. Custom reports. Training beyond the basics. The integrations everyone agreed to defer. All of it gets moved to phase two, and then the partner rolls off to the next project, and phase two never happens.
That’s an unfinished system. The foundation is sound, but the build stopped at the framing. It’s the most common thing we find when a company calls us to look at someone else’s work, and I mean that literally. Across our rescue engagements, unfinished outnumbers broken by a wide margin.
Broken is different, and rarer. Broken means the foundation itself is wrong. The platform genuinely doesn’t fit how the business runs, or decisions made in the first month of the project, like how items and units of measure were structured or how costing was set up, produce numbers that can’t be trusted no matter how much you build on top of them. You can’t configure your way out of broken. You have to rework the foundation.
The distinction matters because the price tags are nowhere near each other. Finishing an unfinished system is measured in weeks. Fixing a broken one can approach the cost of the original project. And here’s the uncomfortable part: a firm quoting your rescue makes far more money on broken than on unfinished. Diagnose this yourself, before anyone with a proposal does it for you.
Five questions that tell you which one you have
- Open the spreadsheets. Take the three Excel files your team can’t close the month without and ask one thing about each: is this doing something the system can’t do, or something nobody ever set up? In most companies, those spreadsheets are rebuilding reports that were scoped, promised, and never delivered. That’s unfinished.
- Trace one bad number all the way down. Pick a single inventory discrepancy and follow it. If the trail ends at a receipt posted three days late or a count done mid-shift with no cutoff, you have a process and discipline gap, which is unfinished. If the trail ends at a unit of measure conversion or an item structure that was set up wrong on day one, you’re looking at the foundation.
- Ask your team why they avoid the system. Listen to the exact words. “It takes too many clicks” and “I never learned that part” are training and configuration problems. “The number it gives me is wrong” is a data problem. The first is unfinished. The second needs a closer look before you can say.
- Find the original scope document. Look at what got moved to phase two during the project. Then look at your list of daily frustrations. If the two lists match, you already have your answer, and you also have the paper trail that proves the work was planned and never done.
- Answer this honestly: knowing what you know now, would you pick this software again? If yes, the platform fits and the job is to finish what was started. If no, and you can name the specific ways it fights how your business operates, that’s a different conversation, and it’s better to have it now than after another year of workarounds.
What the other side looks like
An orthopedic medical supply company came to us convinced they had outgrown their ERP. They distribute surgical kits to hospitals and surgical centers, and the kits stay company-owned inventory until the moment of surgery. By the time they called, roughly one inventory count in four was wrong. Nobody could say with confidence which components were in which kit, where the kits were, or where materials stood in the sterilization and packaging cycle. In their world that goes beyond an accounting problem. A missing component means a procedure could start without the right materials on the table. The teams had built manual workarounds on top of manual workarounds, and leadership assumed the answer was a full replacement.
We did the workflow analysis first, and the analysis said otherwise. The platform could handle their business. It had never been built out to match it. So instead of a replacement, we finished the system they had: lot tracking for full kit and component traceability, an integration that turns surgical usage into replenishment orders automatically, subcontracted workflows to track materials through outside sterilization vendors with quality control on receipt, and the reporting layer leadership had been assembling by hand.
Inventory accuracy went from one in four wrong to effectively zero. More than 800 hours a year of manual effort disappeared, the equivalent of 20 work weeks. And they scaled the operation on the platform they already owned. Any firm quoting that engagement as a replacement would have charged them several times as much to end up in the same place, which is exactly why the diagnosis has to come before the proposal.
Find out which one you have
If this is where you are right now, our No-Cost Technology Assessment was built for exactly this situation. There’s no pitch attached. We look at what happened, tell you whether you’re dealing with broken or unfinished, and lay out what it would take to fix it, in writing, whether you hire us or not. [Book the assessment here]
in your business right now?
