A manufacturing company in the Midwest had just finished an 18-month ERP implementation. The system was solid — real-time inventory, automated purchasing, integrated with their accounting platform. Six months after go-live, the floor supervisors were still running their departments from spreadsheets.
Not because the software was bad. Nobody could find the constraint.
What Most Rollouts Get Wrong
When a new system doesn't stick, the response is almost always the same:
- Get excited about the capabilities
- Announce the go-live date companywide
- Schedule the training sessions
- Launch with a "change management" slide deck
- Watch adoption plateau at 30%
- Add more training
- Wonder why nothing's improving
- Add more features or integrations
- Repeat step 7
It feels productive. Someone's always in a meeting about the system. But nothing ever really moves — and leadership quietly wonders whether the investment was a mistake.
The Turning Point: Theory of Constraints
The frame that changed how I think about this comes from The Goal by Eliyahu Goldratt. It's a book about bottlenecks in manufacturing, but the principle applies perfectly to any software rollout:
If you improve anything that isn't the constraint, you're just adding complexity.
Once you start thinking about adoption as a flow problem — not a training problem — everything shifts.
What Changes When You Look for the Constraint
Instead of asking, "What should we add to the system?"
Start asking: "What is actually stopping someone from completing this task in the system today?" That's the goal.
In most cases, it isn't a missing feature. It's something embedded in a step that already exists, like:
- The login requires a VPN that takes 90 seconds to connect — so people open the spreadsheet instead
- The data entry form has 14 required fields — so supervisors batch-fill it once a week from memory instead of in real time
- The system shows what happened, but not what to do next — so managers don't trust it and make calls without it
After working through enough of these, I've found that every step of the adoption flow matters — but especially the step right before people route around the system. That's your constraint.
Why a Great System Isn't Enough
One operations director we worked with had a genuinely well-designed inventory system — clean interface, accurate data, real results for the customers who used it — but the warehouse team was routing around it constantly. The problem wasn't the software. It was the handoff.
We mapped the exact step where adoption broke: receiving staff had to switch between two screens to confirm a delivery, and the second screen required a separate login they rarely had memorized. That was the constraint — not the system itself.
Fixing that one step — combining the confirmation into one screen with shared credentials — brought compliance from under 40% to over 90% within three weeks.
Not because we added more. Because we fixed the real constraint.
TL;DR: Your team has already told you where the system breaks — in every workaround they've built around it. Find that step, fix it, and don't touch anything else until you do.
