ERP selection: criteria, approach and checklist

A sound ERP selection assesses three things together: the system, the partner and your own starting point. Feature lists rarely decide it, because on paper most systems can do the same. The difference is made by the fit to your own core processes, a partner who demonstrably carries your complexity, a clean migration path for data and customisations, and the costs over five years instead of the quoted price. This page walks through the ten criteria that prove decisive in projects, the approach in five steps and the most common wrong turns with their antidotes.

The ten criteria that really decide

Assess every criterion against verifiable answers instead of demo impressions. Where a vendor evades, that is usually where the later conflict sits.

In DGP projects every criterion is tested with a question the vendor has to answer in the system, not on a slide. Where an answer stays vague, the criterion is recorded as open, not as met.

1. Process fit. Does the standard carry your core processes, order-to-cash and purchase-to-pay, without bending them? Deviations are allowed, but each needs a business reason.
2. The partner’s complexity record. Has the partner demonstrably carried projects of your size and structure, several legal entities, manufacturing depth, countries? References only count if they resemble your situation.
3. Migration path. How do data and customisations get out of the legacy system? A reliable path names test runs, cleansing steps and how the legacy load is handled.
4. Release capability. Does the system stay current without every customisation triggering a new project? In the cloud the version level is standard; what matters is whether your extensions survive the updates.
5. Five-year costs. Licences, implementation, internal effort and operations together, not the quoted price. The internal effort is the most frequently underestimated block.
6. Data model and reporting capability. Do management and the business get the numbers they steer by, without Excel detours? Dimensions and master-data logic decide this early.
7. Ecosystem over point solution. How many extensions, developers and second opinions exist in the market? A broad ecosystem protects against dependence on a single vendor.
8. AI capability on your data foundation. Does AI work inside the system on your data, rather than as a side tool next to it? Over the next years this becomes the difference in daily operations.
9. Responsibilities in the project. Does the vendor name what they need from you, decision-makers, key users, data work? An offer without your obligations to cooperate is calculated too smoothly.
10. Exit and escalation capability. What happens when things stall: clear escalation paths, verifiable interim states, and a contract that does not make a change impossible.

The approach in five steps

1. Describe core processes and the state of the data. Prioritised and with an owner, not as a wish list. What a requirements document contributes and what it does not is covered in Does ERP selection need a requirements document?
2. Narrow the field. Two or three systems that fit your size, instead of ten demos. For the complex mid-market the fundamental question is often Business Central or SAP?
3. Have your own scenarios demonstrated. The vendors show your cases with your sample data, not their standard demo. Only then does the process fit show itself.
4. Vet the partner separately, decide together. How to recognise a partner who carries your complexity is covered in How do I choose an ERP partner?
5. Have the offer cross-read. Before signing, a senior-led second opinion on the offer: days of effort against months of correction inside the project.

Which system fits complex processes?

The honest answer does not start with the system but with the question of how much of your complexity is real and how much is habit. Only then is a system comparison worth it. The full answer is in Which ERP is the right one for us?, and for the upper boundary in Business Central or SAP in the mid-market.

We run this selection as a specialist with Business Central as home, and precisely for that reason the counter-check is part of it: if your situation calls for a different system, we say so in the conversation, before you spend money. Steering stays on your side.

The most common wrong turns

Three patterns appear in almost every selection that goes wrong: deciding on demo impressions instead of your own scenarios, searching for the “best” partner instead of the right one, and a timeline that ignores internal decision-making capacity. All three can be corrected before the start; inside the project they get expensive.

How to tell early whether a consultancy carries the work instead of merely accompanying it is covered in Experiences with Business Central consulting and in What is senior-led ERP consulting?

And if an “independent” selection consultancy is sitting across from you right now: the fair comparison of the two consulting models is in Independent ERP consulting or a specialist: what fits when?

Source: Microsoft Learn: update cycles for Business Central