Where does €1 million hide in a factory?

‍

That question sounds like a gimmick. It isn't. Walk into almost any mid-to-large European manufacturing site that hasn't run a serious energy programme in the last three years, and €1 million of annual waste is sitting somewhere in the building. Sometimes more.

The interesting thing is not the figure. It is how consistently the waste hides in the same four places, regardless of sector, country, or vintage. We see the same pattern in food and beverage plants in the Netherlands, automotive sites in Germany, packaging facilities in the UK, and chemical sites across France and Italy.

The cause is rarely broken equipment, and rarely poor operators. The real problem is that no-one in the building can tell you, in real time, where the energy is going and what it is costing.

This is a visibility problem. And visibility is significantly cheaper than capex.

Here is where to look.

1. The grid layer (€200–400k a year)

Industrial electricity prices in Europe haven't been stable since 2021, and the day-ahead market regularly swings by 10x within a 24-hour window. Most factories run their production schedules in defiance of that volatility. Lines start at 6am because they always have, regardless of whether the day-ahead price for that hour is in the top decile of the year.

Peak demand charges add a second layer. In the UK, DUoS red-band charges between 4 and 7pm can account for 25 to 35 percent of a manufacturer's network costs. Most sites have never quantified how much of that is structurally avoidable.

The third grid-layer leak is capacity market exposure. UK Capacity Market obligations and their continental equivalents quietly add several pounds per kW of peak demand to the bill. Sites that flex their peak by even 10 percent see meaningful reductions, and most have never tried.

These three lines on the bill, taken together, comfortably reach €200,000 to €400,000 a year for a site spending €2m or more on electricity.

2. The asset layer (€250–400k a year)

Compressed air is the textbook example. The rule of thumb is that every 1 bar reduction in system pressure cuts compressor energy consumption by around 7 percent. Walk a typical factory and you will find the system running at 7.5 bar because that is where it was set when the third compressor was installed in 2017, even though the actual line demand peaks at 6 bar.

Chillers and refrigeration are similar. Setpoints get tightened during a heatwave one summer and never get loosened again. A 1°C lift on a chiller setpoint is roughly 3 percent off its energy bill. Most sites have several degrees of headroom they have never reclaimed.

Motor drift is the quietest leak. Motors degrade slowly. Bearings wear and alignments shift, while drives lose efficiency before anyone notices. Without continuous monitoring, the first signal is usually the motor failing. By that point it has been wasting energy for months.

Across compressed air, refrigeration, and motor systems, the typical recoverable spend is €250,000 to €400,000 a year on a mid-sized site.

3. The waste layer (€150–300k a year)

This is the cheapest leak to find and almost always the most embarrassing.

Pull your site's half-hourly consumption data for last Sunday at 3am. Compare it with your weekday production peak. If your weekend overnight low is more than 30 percent of your peak, you have a baseload problem. Best-in-class sites run below 15 percent.

What is drawing electricity at 3am on a Sunday? Compressors cycling against air leaks. Extract fans nobody can remember commissioning. Refrigeration loads idling in unused storage. Office HVAC running a weekday schedule because no-one updated the BMS calendar after the last bank holiday.

The cost is straightforward to calculate. Each percentage point of unnecessary baseload represents roughly 4,500 hours a year of wasted electricity. On a €2m electricity spend, that is around €20,000 per percentage point. Most sites we look at have between 8 and 15 percentage points to recover.

4. The commercial layer (€100–200k a year)

This one is on the bill, but no-one reads the bill closely enough to find it.

Reactive power penalties are the first item. Sites with poor power factor pay surcharges that can run into tens of thousands a year, and the fix is often a one-off correction project that pays back inside 18 months.

Maximum Import Capacity is the second. Many sites are paying standing charges on capacity they no longer need. A peak demand reduction programme often shrinks the required MIC by 10 to 20 percent, and that is a permanent reduction in the standing charge for as long as the site operates.

Flex revenue is the third. UK and European DSR markets, capacity market participation, and increasingly the equivalents of the old Triad windows all pay industrial sites for being responsive. Sites that participate properly recover £15–30 per kW of flexible load per year. Most don't participate at all.

Visibility is cheaper than capex

The pattern across all four layers is the same. The €1m of waste isn't there because the equipment is wrong. It is there because no-one is watching closely enough to spot it, in real time, against the right prices.

Most sites already have the meters and the BMS. The half-hourly data is usually sitting in a supplier portal already. What they don't have is the operating model that converts that data into decisions, every day, on every shift.

That is what predictive optimisation is for. Not replacing the equipment. Making the equipment already there visible enough to be managed properly.

Where to start

If your site spends more than €1m a year on energy and you can't tell me, this afternoon, where the off-shift baseload comes from and what it costs you in a year, that is the cheapest place to start.

The first audit doesn't need new hardware. It needs a week of half-hourly data, an honest look at the bill, and someone willing to ask the embarrassing questions.

The €1 million isn't going anywhere until you do.

‍