Time-to-Live OEE: Why It's the Only MES Selection Criterion That Matters in 2026
Most MES evaluations focus on features: “Does it do SPC? Does it have scheduling? Does it integrate with SAP?” These are the wrong questions — or at least, not the first question.
The first question is: how long until I have OEE data from my first machine?
Because if the answer is “3–6 months,” everything else in the evaluation is moot. You won’t see ROI in the project timeline your board approved. Your operations team will lose confidence in the rollout before the first KPI is live. And you’ll spend the first year managing an implementation project instead of improving production.
Why most MES rollouts fail at time-to-live
The implementation time problem in MES comes from four sources:
1. Machine connectivity is the critical path. If you can’t get data off the machine, you can’t show OEE. Most legacy MES implementations assume a dedicated OT/IT team and months of custom integration work for each machine type. A process that requires a site visit, a custom driver, and 6 weeks of testing per machine doesn’t scale.
2. Change management takes longer than software deployment. Operators have to change how they work. Supervisors have to change how they manage. That’s a people problem, not a software problem — but it’s still the implementation team’s problem. Tools that require significant training and behavior change take longer to adopt.
3. Pilot scope creep. “Let’s start with 3 machines” turns into “we need to configure 47 items, 200 downtime codes, and 15 work centers before we can go live.” The right answer is to get the first machine live with good OEE data in the first week, then expand. Most MES tools don’t support a narrow-scope-first approach.
4. Data quality problems appear late. If machine connectivity is imprecise, OEE data is wrong. Wrong OEE data is worse than no data — it creates disputes between operators and supervisors, undermines confidence in the system, and delays full rollout by months while teams argue about whether the numbers are real.
What a fast time-to-live looks like
For OPC-UA connected machines, a well-designed MES should have live OEE data within 2 hours of initial configuration — not 2 weeks. That means:
- Browse the machine’s OPC-UA address space from the MES UI
- Map the production count and fault signals to OEE inputs
- Define basic downtime categories (planned / unplanned / changeover)
- Start collecting data
No custom driver development. No on-site vendor visit. No 6-week testing cycle.
For machines without OPC-UA (older CNC, injection molding, assembly equipment), MQTT and pulse-counter connections should extend the same fast-start model to any machine with an electrical output signal.
How to evaluate vendors on time-to-live
Ask these questions in the demo:
“Can you connect to a machine in this demo environment right now?” If the vendor needs to prepare a demo environment in advance and can’t show live machine connection, that’s a signal.
“What is your fastest documented customer go-live?” Get a specific example, not a marketing claim. Fastest go-live for OEE data, not full system rollout.
“What does week 1 look like?” A credible answer is: one machine connected, live OEE on that machine, first downtime event categorized. An incredible answer is: “we spend week 1 on requirements gathering.”
“What does your edge hardware requirement look like?” Some MES vendors require dedicated edge servers or proprietary hardware. Others run connectivity software on existing infrastructure. Hardware procurement adds weeks.
The feature breadth trap
The vendors with the longest feature lists are usually the vendors with the longest implementations. Enterprise MES platforms (Siemens Opcenter, Rockwell FactoryTalk, SAP ME) have comprehensive feature sets that took 20 years to build. They also routinely take 12–18 months to deploy.
For most manufacturers, 80% of the OEE value comes from 20% of the feature set. Getting that 20% live in week 2 creates more business impact than getting 100% live in month 18.
The right evaluation sequence: live time-to-OEE data first, then feature depth second. If a vendor can’t get you data in week 1, their feature list is irrelevant.
See Qontiv’s Edge Gateway for how machine connectivity works. See OEE & Analytics for what live OEE looks like once you’re connected. If you want to see a live machine connection in a demo, request one — we’ll connect to a simulated machine during the session.