The ERP selection process at most mid-market manufacturers is broken in a way that is almost perfectly designed to produce bad outcomes. It is led by the wrong people, evaluating the wrong things, against criteria that have no relationship to what will determine whether the implementation succeeds or fails. The result is a predictable pattern: a system is selected based on a demonstration it cannot replicate in production, implemented by a team that underestimated the complexity, and declared a success before anyone has honestly assessed whether it is delivering the value that justified the investment.
What Gets Evaluated and What Should
The typical ERP selection process is dominated by feature comparison. Vendors demonstrate capabilities against a requirements list. Evaluators score each vendor on coverage. The vendor with the highest coverage score wins. This process optimizes for the wrong outcome. Feature coverage tells you nothing about whether a system will work in your specific operational context, whether it will be adopted by the people who have to use it, or whether the implementation partner has the capability to close the gap between what the software can do and what your business actually needs.
The questions that actually predict implementation success are rarely asked in formal selection processes. How does this system handle the specific edge cases in your production environment — the jobs that don’t fit neatly into the standard work order model, the materials that don’t behave like the system expects? What does the upgrade path look like, and who bears the cost and risk? How does the implementation partner typically handle scope changes when the reality of your operation doesn’t match the assumptions in the implementation plan?
Feature coverage tells you nothing about whether a system will work in your specific operational context, whether it will be adopted by the people who have to use it, or whether the implementation partner can close the gap.
— Industrial Foresight Analysis, 2026
The Partner Problem
The implementation partner decision is at least as important as the software decision, and it receives a fraction of the scrutiny. Manufacturers spend months evaluating ERP vendors and hours evaluating implementation partners, when the evidence consistently shows that implementation quality is the primary determinant of whether an ERP project delivers its promised value. A capable implementation partner can make a mediocre software choice work. An inadequate implementation partner will fail with the best software in the market.
The evaluation of implementation partners suffers from the same feature-coverage bias as the software evaluation. References are called, methodology decks are reviewed, team CVs are inspected. The one thing that is almost never evaluated is the partner’s honest track record — what percentage of their implementations come in on time and within the original budget, what their customers’ actual adoption rates look like eighteen months after go-live, and whether the relationship between partner and client tends to be collaborative or adversarial when things get difficult.
The Customisation Trap
Mid-market manufacturers have a particular vulnerability to the customisation trap. The instinct to configure software to match existing processes rather than adapting processes to leverage software capabilities is nearly universal, and it is nearly always the wrong choice. Every customisation is a debt — it must be maintained through upgrades, it creates dependency on the original developer, and it adds complexity to every future change. The customisations that feel essential during implementation frequently feel like millstones two years later.
The manufacturers who emerge from ERP implementations in the strongest position are almost always the ones who accepted significant process change as part of the project. They used the implementation as an opportunity to standardise, simplify, and rationalise operations that had accumulated decades of informal workarounds. They ended up with systems they could actually maintain and evolve, rather than highly customised implementations that only their original implementation partner fully understands.
A Better Selection Process
The selection process that produces better outcomes starts differently. It begins with an honest assessment of the organisation’s process maturity and change management capability, because these factors constrain what any ERP implementation can achieve. It identifies the two or three operational outcomes that would make the investment clearly worthwhile, and it evaluates vendors and partners against their demonstrated ability to deliver those outcomes in comparable environments.
It includes customer conversations that go beyond the reference list — conversations with customers the vendor did not suggest, at a point in the relationship where the honeymoon period has ended and the real texture of the system and the partnership are visible. And it treats the implementation partner selection with the same rigour as the software selection, because the evidence consistently shows that partner quality is the variable that matters most.
Panorama Consulting — 2026 ERP Report