What should a manufacturer look for in an ERP?

Fit with how you actually manufacture, honest integration with what happens on the floor, and a realistic view of implementation effort. The platform comparison is the easy part; the difference between a successful ERP and an expensive one is almost entirely in the second and third.

For Australian manufacturers of this size the field usually narrows to a mid-market platform such as MYOB Advanced or a comparable product, sometimes an industry-specific system built for a particular kind of production. Both routes work. The general platform brings a larger ecosystem and more people who know it; the specific one arrives understanding your process. Weigh the second consideration carefully, because configuring a general system into understanding your manufacturing is where budgets go.

Fit means matching your actual process rather than a tidied version of it. Discrete assembly, process manufacturing, make-to-order and make-to-stock have genuinely different requirements, and a platform strong in one can be awkward in another. So does the complexity of your bills of materials, whether you need lot or serial traceability, and whether production scheduling is a real constraint or a light overlay. Trial with your own products and your awkward cases, not with the vendor's demonstration data, because every ERP looks capable on clean examples.

Integration with the floor is where expectations most often part company with reality. Ask a specific question about your specific equipment: can this platform receive production data from these machines, and has the vendor done it before with this equipment. General claims about connectivity are not answers. Where native integration does not exist the options are middleware, a manual data entry step, or accepting a gap, and all three are workable provided the decision is made knowingly rather than discovered after go-live.

Implementation effort is systematically underestimated, and the reason is consistent: the hard part is data, and data belongs to the business rather than the vendor. Product records, bills of materials, routings, supplier details and inventory all need to be accurate before migration, and most manufacturers discover during this exercise that their existing data is less clean than assumed. That clean-up is genuinely useful work and it is work, done by people who also have day jobs, and no vendor timeline survives contact with it.

The infrastructure questions are simpler and worth settling early. Whether the platform is cloud-hosted or needs a server, what connectivity it requires, how it behaves for staff on the floor using terminals or tablets in a harsh environment, and how it performs across sites if you have more than one. Shop-floor access in particular deserves attention, because a system designed for office desks does not necessarily work well on a screen next to a machine with someone wearing gloves.

Then the questions that protect you later. Who owns your data and in what format can it be exported. What does support cost and what does it cover. What happens at renewal. How are upgrades handled and can they be deferred, since an upgrade forced during a busy production period is exactly what nobody wants. A confident vendor answers all of these plainly, and vagueness on any of them is information.

Plan the go-live around production rather than the vendor's schedule, and expect a dip. Even a well-run ERP implementation costs productivity while people learn the tool they now use for everything, and doing that during a peak or a major order is a decision the business will regret. Keep the old system readable for a defined period, verify migrated data rather than counting records, and give people a second training session a fortnight in when they have real questions.

One organisational point that outlasts the platform choice: name an internal owner. An ERP with no owner drifts out of configuration, workarounds accumulate, and within two years the business is running on spreadsheets alongside a system it is still paying for. That role does not have to be technical and it does have to be accountable, which is what the vendor and your provider both need on the other end of the phone.

The honest caveats. ERP projects have a poor reputation for good reasons, and the failures are almost always about scope, data and ownership rather than software. A smaller, well-implemented system beats an ambitious one that is half-configured. And if your current platform is disliked, establish whether the problem is the software or the implementation before replacing it, because reimplementation is often cheaper. If you want the infrastructure and integration side assessed, call 1800 456 567.