What Decades of ERP Lessons Teach Us – Your Practical Manufacturers Guide to AI
Why Technology Fails in Manufacturing
- processes weren’t defined
- leadership wasn’t aligned
- the focus was on features, not results
- lack of agreement on what defined success
- What exact problem are we trying to solve?
- Why do we think AI is the right tool?
- What specific result do we expect to see?
Start Simple: A Business Case, One Page, Clear Goals
- The exact problem
- The process gaps
- The measurable result expected
- The financial impact (ROI)
A Successful Business Case: Combilift, Global Forklift Manufacturer
A great example comes from Combilift, the global forklift manufacturer. They didn’t use AI because it was trendy. They wanted to use it because they found real process problems:
New parts staff were missing revenue
Quotes were incomplete or inaccurate
Customers were frustrated
So they paired Infor AI with Infor CloudSuite Industrial to guide parts selection and fix the issue. Within 60 days, they realized:
30% increase in first-time fixes
40% reduction in service job costs
30% increase in revenue per transaction
Combilift case study is what happens when AI is used to solve a specific, well-defined problem — not a vague “digital transformation” goal. Tools like process mining make the problem definition phase easier by showing you:
Where processes get stuck
Where mistakes happen
What parts of the process are not following standard flow
It’s the same logic used in ERP success:
define the problem → understand the process → fix the root cause.
Trust Proof
Manufacturers have learned the hard way a polished demo doesn’t prove anything. AI is no different. A generic, canned presentation can make any solution look impressive, but it doesn’t tell you whether the tool can support the realities of your plant, your workflows, or your constraints.
The safer approach is to judge AI solutions based on how well they align with your actual processes. Before being impressed by screens or automation, manufacturers should take the time to:
Map out how work actually moves through their operation
Identify where delays, rework, and inconsistencies appear
Clarify why each process needs to change
Define which outcomes matter most
With this clarity, you can challenge vendors to show how their solution addresses your specific process gaps, not an idealized version shown in a controlled environment. You don’t need full access to your data or a full implementation to do this, what you need is a demonstration that speaks directly to your workflows and problem areas.
Manufacturers who evaluate AI this way quickly separate practical solutions from marketing theater.
A Simple Chart to Overcoming the Common Bariers
| Barriers to Start | What do you need to do? |
| Not knowing where to start | Diagnose processes first; build a business case |
| Overwhelmed by technology updates | Focus on critical needs, not feature richness |
| Lack of IT resources and knowledge | Use expert-implemented solutions; don't over-customize |
Conclusion: Trust Evidence, Not Marketing
- how information moves between teams
- where delays or errors occur
- what triggers rework
- and why a process needs to change in the first place
How Possible Happens - Fill the 'value-void' between what technology promise and output
Discover how manufacturers like you are cutting wasted hours, tightening cost control, and streamlining reporting with real, measurable improvements.
Infor Velocity Suite: Streamline your processes and enhance efficiency with Generative AI, RPA, and Process Mining
Package of solutions and services that makes process innovation easy and impactful. Using RPA, plus GenAI and Process Mining – along with implementation services.
