Free OEE POC: What a Real Proof of Concept Looks Like, How 48 Hours Produces Decisive Data, and Why Manufacturers Choose This Path Over Trials

A free OEE POC — proof of concept — is the fastest path from uncertainty about production performance to decisive data you can act on. Unlike a software trial, which gives you access to a platform and asks you to figure out how to use it, a proof of concept is a structured deployment on your actual production lines, with real sensors, real data, and a clear outcome: at the end of 48 hours, you know your real OEE including the micro-stops that manual measurement never captures, you see the root causes of your losses ranked by impact, and you have a quantified improvement potential figure for each of your lines. This is what a well-run OEE proof of concept produces. This guide explains exactly what happens during those 48 hours, what you should expect to see, and how to prepare so the outcome is maximally useful.

Before going further, a preliminary note: if you are still in the awareness phase — trying to understand what OEE is, how it is calculated, what a reasonable target looks like — a POC is probably premature. The right starting point is the free OEE Excel calculator, which produces a first baseline in 15 minutes and builds the business case for what comes next. A POC is the right next step once you have that baseline and want to see what the reality beneath the manual measurement actually looks like.

What a free OEE POC actually is — and what it is not

A free OEE proof of concept is the installation of a full OEE measurement platform on one or more of your actual production lines, for a defined period (typically 48 hours of active production), at no cost to you and with no purchase commitment at the end. Non-intrusive IoT current sensors clip onto each machine’s power supply cable in 10 minutes per machine, without stopping production, without modifying equipment, and without requiring any IT integration. A Field V4 operator tablet is positioned at the line for downtime reason tagging. The cloud platform begins ingesting data the moment sensors are connected.

What a POC is not: it is not a sales demonstration, it is not a theoretical presentation with slides, and it is not a platform access trial where you are expected to deploy hardware yourself. TeepTrak’s POC program is structured around the deployment engineer doing the installation on-site (or remote-guided for simpler setups), the platform being tuned to your specific machine types and product references, and the analysis of the resulting data being delivered as a written report alongside live dashboard access. The manufacturer’s time investment is approximately 2 hours total — the sensor installation and a 30-minute review at the end. The vendor does the work; you receive the answer.

The 48 hours, step by step

The first 2 hours are sensor installation. Current sensors clip onto the power supply cables of each machine in the POC scope — typically 3 to 6 machines for a meaningful sample. The edge gateway that aggregates sensor data is positioned near the line and connected to LTE or industrial WiFi independently of the factory IT network (no firewall changes, no IT approval required). The operator tablet is configured with your product references and downtime reason categories. Production continues normally throughout this installation.

The next 46 hours are normal production, measured automatically. Every machine cycle is detected. Every stoppage is timestamped to the second. Micro-stoppages under 5 minutes — the events that manual tracking never captures — are recorded alongside major events. Operators tag the reasons for major stoppages on the tablet, following the categories you defined. The cloud platform aggregates all of this into real-time OEE calculations for Availability, Performance and Quality at the line, shift, machine and product reference levels. No one from your team needs to do anything during these 46 hours; the platform measures continuously while normal production runs.

At the end of 48 hours, the analysis produces a complete picture: your actual OEE across the sampling period, decomposed into the three standard components; your Pareto of loss categories ranked by total time impact; the cross-dimensional patterns that JEMBA AI identified — which losses correlate with specific shifts, products, operators or machine conditions; and the quantified improvement potential for each line, expressed in both OEE percentage points and annual production value in euros. This is the decisive data the POC is designed to produce.

What most manufacturers discover during the POC

The specific insight that a 48-hour automated POC produces — and that no manual measurement can produce, regardless of how carefully it is executed — is the gap between perceived and actual OEE. Manufacturers who have been tracking OEE in Excel for months typically enter the POC expecting a number similar to what their manual calculation has shown. The reality is almost always 10 to 25 percentage points lower, because the manual calculation was missing the micro-stops, the brief speed reductions, and the quality events that were never logged.

The reaction in the review meeting at the end of the POC follows a consistent pattern: initial surprise that the real number is so different, followed by recognition that the Pareto of losses now makes sense of patterns the team had observed but never quantified, followed by the realisation that the improvement opportunity is significantly larger than previously estimated. At a typical line running 4,500 hours per year at 200 euros per hour of added value, each OEE percentage point is worth approximately 9,000 euros annually. A POC that reveals 15 hidden OEE points per line translates into 135,000 euros per line per year of previously invisible improvement opportunity. For 4 production lines, 540,000 euros annually.

This is why the POC produces decisive data. It is not a matter of whether to improve OEE — that was always going to be the case. It is a matter of knowing accurately where the improvement opportunities actually are, so the investment and effort go to the right places. The OEE software pricing and ROI guide covers the financial framework in detail.

How to prepare for a POC that maximises useful outcomes

Three preparation steps make the difference between a POC that produces useful data and one that produces ambiguous data. First: identify the right lines. The ideal POC scope is 3 to 6 machines on lines representative of your production mix — not the best performers and not the worst performers, but the lines where improvement decisions will be made. Including lines at both extremes produces noise; including representative lines produces insight.

Second: align the timing with normal production. A POC during an unusual production week — seasonal ramp, new product launch, planned maintenance period — produces data that does not reflect normal operation. Schedule the POC for a period that represents what your operation looks like most of the time. Third: ensure the product references running during the POC cover your typical mix. If 70% of your normal production runs Product A and 30% runs Product B, the POC period should roughly mirror that distribution so the data is representative.

The free OEE Excel calculator we provide is useful preparation for the POC conversation. Fill it in for one pilot line before the POC starts, so you have a manual baseline to compare against the automated measurement. The gap between the two is the most valuable data point the POC produces, because it quantifies the invisible loss you have been living with.

What happens after the 48 hours

Two outcomes are possible at the end of a POC, and both are legitimate. The first: the data confirms that your current OEE is close to what you expected, the micro-stops are not a major factor, and your existing measurement is accurate enough to drive your improvement decisions. This is a genuine outcome — you have confirmed that your investment in automated measurement would not produce significant additional insight. We tell manufacturers this honestly when the data supports it. The POC has still been valuable because it has answered the question definitively rather than leaving it as an assumption.

The second outcome, which is what happens in the overwhelming majority of POCs: the data reveals significant invisible losses that change the improvement priorities and justify the investment in automated measurement. At that point, the commercial discussion becomes straightforward — the SaaS pricing of the platform is compared to the quantified annual improvement opportunity, and the ROI calculation is built from POC data rather than vendor estimates. Manufacturers who make commercial decisions based on POC data consistently recover the first year of platform cost within the first 4 to 8 weeks of deployment, because the improvement is based on the real Pareto rather than the manual one. For the complete platform details, see our OEE software guide and the production monitoring software overview.

Start your free 48-hour POC today — on your actual production lines
IoT sensors installed by our team · real OEE data · JEMBA AI root cause · no commercial commitment
Request your free TeepTrak POC

External references: MESA International — OEE research and benchmarks · Wikipedia: Overall Equipment Effectiveness

See also: Free OEE software guide · Free OEE calculator Excel template · Free OEE software trial · OEE software complete guide