MES Selection Criteria: 8-Point Evaluation Framework 2026

TL;DR: MES selection criteria determine whether a Manufacturing Execution System delivers measurable business value — or turns into a long IT project. The 8 criteria below are not a theoretical framework. They're distilled from 15,000+ machine connections and hundreds of vendor evaluations. Each criterion includes what good looks like, red flags to watch for, and the proof question to ask every vendor. At the bottom: a scoring matrix you can adapt to your steering committee.

Table of contents

1. What are the 8 MES selection criteria that actually matter?
2. Criterion 1: Time-to-value
3. Criterion 2: Economic focus over feature depth
4. Criterion 3: Machine connectivity & integration
5. Criterion 4: Cloud vs. on-premise deployment
6. Criterion 5: Scalability & multi-plant capability
7. Criterion 6: Usability & shop-floor adoption
8. Criterion 7: Proof of Value before commitment
9. Criterion 8: Total cost & ROI
10. Scoring matrix: The evaluation framework you can reuse
11. FAQ

What are the 8 MES selection criteria that actually matter?

Most MES selection guides list 20–30 criteria and treat them equally. In practice, 8 criteria account for 90 % of the difference between a successful implementation and a failed project. The table below shows each criterion, what to look for, and the red flags that predict failure.

Criterion 1: How fast does the MES deliver reliable KPIs?

Time-to-value is the single most predictive criterion for MES success. Systems with long implementation phases delay ROI, erode internal support, and increase the risk of project cancellation. The question isn't "how many features does it have?" but "how quickly does it show me what I'm losing?"

Customer benchmark: Klocke (pharma packaging) went from first machine connection to full plant coverage in 3 weeks. Meleghy (automotive) scaled from pilot plant to 6 plants in 6 months. At Neoperl, a 4-week PoC on one machine validated the business case — contract followed.

Criterion 2: Does the MES quantify losses in euros, not just KPIs?

An MES must make production losses visible — but visibility alone doesn't drive action. What drives action is knowing that machine 7 cost € 23,000 in avoidable downtime last month, and the top cause is changeover overruns. Without cost context, even detailed OEE data results in a weak business case and no improvement actions.

What to look for: Clear attribution of downtime, scrap, and throughput losses to € values (€/hour, €/unit, €/shift). Pareto ranking by financial impact, not just frequency. Automatic ROI tracking showing how much has been recovered.

Criterion 3: How does the MES connect to your machines?

Connectivity determines project cost and scalability more than any other criterion. If every machine type requires custom development, your "MES project" is actually an integration project — and integration projects never end.

🚩 Red flags: Proprietary drivers that lock you into one vendor. Custom development per machine type. Dependency on an external system integrator for every new connection. If connecting 10 machines costs more than the first year of MES subscription, the connectivity model is wrong.

Criterion 4: Should you choose cloud or on-premise MES?

The deployment model directly affects TCO, maintenance effort, and scalability. This is covered in depth in the Cloud MES vs. On-Premise comparison. For evaluation purposes, the key trade-offs:

Criterion 5: Can the MES scale from pilot to multi-plant?

Many MES solutions work in the pilot — and fail during rollout. The evaluation must test scalability before you commit, not after.

What to look for: Can the first plant's configuration be replicated to a second plant without starting over? Are KPI definitions standardized or plant-specific? Can the customer's own team expand the system, or is the vendor needed for every change?

Customer benchmarks: Meleghy: 1 pilot plant (Wilnsdorf) → 6 plants across 4 countries in 6 months. Carcoustics: replaced legacy MES, scaled to 500+ machines in 7 countries in 6 months. Both teams expanded SYMESTIC independently using the modular platform — no vendor dependency for new connections.

Criterion 6: Will the shop floor actually use the MES?

Usage determines success. The most feature-rich MES is worthless if operators don't use it and supervisors work around it. Low adoption is the #1 killer of MES projects — ahead of technical issues, connectivity problems, or cost overruns.

What to evaluate:

• Can an operator use it productively within 30 minutes without formal training?
• Are there role-based views (operator, supervisor, management, maintenance)?
• Is there mobile access for leadership (smartphone/tablet)?
• Does the vendor offer a key user onboarding (8–12 hours) that enables self-sufficiency?

Criterion 7: Can you prove the value before you commit?

Modern MES selection is based on evidence, not promises. A vendor who can't (or won't) run a pilot on your real machines is a vendor who isn't confident in their own product.

What a good Proof of Value looks like:

• Time-boxed: 30 days
• On your real machines with real production data
• Predefined target KPIs agreed upfront
• Clear go/no-go decision at the end
• Zero contractual commitment required

Customer proof — Neoperl: 4-week PoC on one assembly machine. Validated functionality and calculated ROI from real data. After successful PoC: contract and integration of first 3 machines. Since then: continuous expansion. The PoC model reduces investment risk and accelerates internal approvals — especially when internal skepticism is high.

Criterion 8: What is the total cost — and how fast does it pay back?

Licence fees alone are misleading. The best MES is not the one with the lowest licence price — it's the one with the fastest ROI. For full cost analysis, see What should an MES cost in 2026? and MES ROI: Business Case Calculator.

What to include in TCO:

• SaaS subscription or licence costs
• Implementation and onboarding (vendor side)
• Internal effort (key users, project coordination)
• ERP integration costs (ABAP development for SAP, file interfaces for others)
• Ongoing operations (server admin for on-premise, nothing for Cloud)
• Scaling costs per additional plant
• Hidden costs: upgrade projects, system integrator dependency, version-lock

Scoring matrix: The evaluation framework you can reuse

Use this matrix to score each MES vendor during your evaluation. Weight the criteria based on your specific priorities. The suggested weights reflect what matters most in practice for mid-market discrete manufacturing.

How to use this matrix: Score each vendor on each criterion (1 = poor, 5 = excellent). Multiply by weight. Sum the weighted scores. Compare vendors on the total. A score > 4.0 indicates strong fit. A score < 3.0 on any single criterion with weight ≥ 15 % is a disqualifier.

Practitioner tip: Don't let feature checklists dominate the evaluation. A vendor with 200 features and 18-month implementation scores worse than a vendor with 50 features and 30-day go-live — because the features you never use cost money, and the 12 months you wait for value erode the business case. Weight time-to-value highest (20 %) for this reason.

FAQ

What's the single most important MES selection criterion?
Time-to-value. How quickly does the system show you what you're losing? If it takes 6 months to see the first dashboard, you've already lost the business case. If it takes 30 days, the system proves itself before internal skepticism can kill the project. Weight this criterion highest.

Should we create a long feature checklist for MES evaluation?
No. Feature checklists favour incumbent enterprise vendors with 200+ features — most of which you'll never use. Instead, focus on the 8 criteria above and ask for proof on each one: "Show me a customer who achieved this." Named references are worth more than feature checkboxes. For a vendor feature comparison, see the MES Software article.

How many vendors should we shortlist?
3–5 for initial evaluation. 2–3 for Proof of Value. 1 for contract. More than 5 makes the evaluation process longer than the implementation — which is itself a red flag. The shortlist should include at least one cloud-native vendor and one traditional on-premise vendor to test the architectural assumption.

Who should be on the MES evaluation team?
Production lead (owns the outcome), IT lead (validates integration), CI/OpEx manager (defines KPIs), and one shift supervisor (validates usability). If IT leads alone, you get an IT project. If production leads, you get a production improvement. MES is a production tool.

How do these criteria apply to small manufacturers (< 50 employees)?
Most of them still apply, but weight differently. Small manufacturers should weight time-to-value (20 %) and total cost (20 %) highest. Scalability and multi-plant are less relevant. A starter package with production KPIs (no full MES) may be the right entry point.

The key takeaway: MES selection is not about finding the system with the longest feature list. It's about finding the system that delivers measurable production improvements fastest, at predictable cost, with proof you can verify. The 8 criteria above — with their red flags and proof questions — give you the framework. The scoring matrix lets you present the evaluation to your steering committee. Use them.

→ What is MES? · → MES Software Compared · → Cloud vs. On-Premise · → MES Costs · → MES ROI · → Best MES System