Signs your maintenance strategy is costing you more than you think

Here’s an uncomfortable question. If your most critical machine failed tomorrow morning, could you tell us, roughly, what that would cost per hour?

Most plant managers can’t. Not because they’re bad at their jobs, but because the true cost of unplanned downtime hides in a dozen different budget lines. Overtime here. A scrapped batch there. An emergency courier charge for a bearing that should have been on the shelf.

And that’s the trap. A maintenance strategy can look perfectly healthy on paper while quietly bleeding money underneath. So let’s pull the floorboards up. Here are the signs that a machine shutdown is costing you money right now, even if it hasn’t happened yet.

First, what does unplanned downtime actually cost?

Before we get to the warning signs, let’s put a number on the problem, because Google will happily give you scary headline figures, and most of them undersell the reality for process industries.

The commonly quoted benchmarks put unplanned downtime at anywhere from a few thousand pounds per hour for a small manufacturer to well over £100,000 per hour for large automotive, oil and gas, or continuous process plants. But the hourly rate is only the visible part. The full cost of a failure includes:

• Lost production that you may never claw back if you’re running at capacity
• Labour costs for staff standing idle, plus overtime to catch up
• Repair premiums, because emergency callouts and expedited parts cost far more than planned work
• Scrapped or reworked product, especially in batch processes where a mid-run stop ruins everything in the vessel
• Secondary damage, where one failed component takes out the machine around it
• Knock-on costs like missed delivery penalties, damaged customer relationships and, in the worst cases, safety and environmental incidents

Add those together and a “two-hour breakdown” often costs five to ten times the lost production alone. If you want to model this for your own site, Faraday Predictive’s benefits calculator will do the sums with your numbers rather than industry averages.

Now, the warning signs.

Sign 1: Your maintenance budget is mostly spent reacting

Look at where your maintenance hours actually go. If more than a quarter of them are spent on breakdowns rather than planned work, your strategy is running you, not the other way round.

Reactive maintenance costs roughly three to five times more than the same job done on a planned basis. The work itself is rushed, done at antisocial hours, and often on a machine that has damaged itself further by running to failure.

Think of it like a toothache. A check-up costs you an hour and a small fee. Ignore it, and the same tooth eventually costs you an emergency root canal, a crown, and a week of misery. The tooth was always going to need attention. You just chose the expensive version by waiting.

If this sounds familiar, our guide on how to build a business case for predictive maintenance shows how to turn that reactive spend into an argument your finance director will actually listen to.

Sign 2: You maintain machines on a calendar, not on condition

Time-based maintenance feels responsible. Every six months, strip the pump, replace the seals, tick the box.

Here’s the problem. Studies going back to the airline industry in the 1970s have shown that most equipment failures are not age-related. Machines don’t fail because a date arrived. They fail because of misalignment, imbalance, bearing wear, cavitation, electrical faults, and other conditions that develop on their own schedule.

So calendar-based maintenance costs you twice:

• You pay to service healthy machines that didn’t need touching, and every intrusive strip-down carries its own risk of introducing a fault
• You still get failures in between services, because the calendar can’t see a bearing that started degrading last Tuesday

The alternative is maintaining machines based on their actual condition. If you’re not sure how that differs from the buzzwords, this explainer on condition monitoring vs predictive maintenance breaks it down clearly.

Sign 3: Failures always seem to come as a surprise

“It just went.” No warning, no symptoms, nothing.

Except that’s almost never true. Rotating machinery rarely fails without notice. Bearings, motors, pumps, and gearboxes typically show measurable signs of distress for weeks or months before they let go. Changes in vibration signatures, in the electrical signals flowing through the motor, in energy consumption.

The failure wasn’t sudden. It was just invisible to you.

This is exactly what techniques like electrical signature analysis (ESA) are built for. By analysing the motor’s own electrical supply, you can detect developing faults in the motor and the equipment it drives, without fitting sensors to hard-to-reach machinery. The motor becomes its own sensor.

A quick real-world example

Picture a water utility running a critical borehole pump. It passes every visual inspection. Then one Friday night it seizes, floods the site with emergency callouts, and takes four days to replace because the spare is on a six-week lead time.

A condition monitoring system watching that pump’s electrical signature would have flagged the developing bearing fault weeks earlier. Same repair, but done on a Tuesday morning, with the part in hand, at planned-work prices, with zero lost supply. That’s the whole difference between a line item and a crisis.

Sign 4: Your “spare motor” strategy is your insurance policy

Plenty of sites cope with unreliability by holding spares. A shelf of motors, a spare pump, a rewound gearbox in the stores.

It works, sort of. But look at what it costs:

• Capital tied up in assets doing nothing
• Storage, insurance and maintenance of the spares themselves (yes, stored motors degrade too)
• It doesn’t prevent the downtime, it just shortens it. You still stop, still swap, still lose production

Spares are a good backstop. They’re a terrible strategy. If your stores are your main defence against a machine shutdown costing you money, you’re paying for insurance against a fire you could simply stop lighting.

Sign 5: Nobody knows the health of your critical assets right now

Ask a simple question at your next meeting: “Which of our top ten critical machines is in the worst condition today?”

If the honest answer is a shrug, that’s a cost. Not knowing means every planning decision, every shutdown scope, every spares order is a guess. You’ll over-maintain some machines, under-maintain others, and get blindsided by the one nobody was worried about.

Modern monitoring makes this a solvable problem, even on a budget. A permanently installed system like Faraday’s Inpod condition and energy monitoring system sits in the switchroom, not on the machine, so it can watch assets that are remote, submerged, or hazardous to access. We also offer the P100 portable equipment health assessor, which lets your own team walk the plant and health-check machines one by one.

And if you’re worried your older equipment is “too legacy” for this, it almost certainly isn’t. Here’s why: Predictive maintenance for legacy equipment: is it possible?

Sign 6: Your energy bills are quietly rising

This one surprises people. A machine that is developing a fault usually becomes less efficient before it fails. Misalignment, imbalance, and bearing wear all waste energy as heat, noise, and vibration.

So a degrading maintenance strategy doesn’t just show up as breakdowns. It shows up on your electricity bill, every single month, across every slightly unhealthy machine on site. Monitoring systems that track both condition and energy, like the Inpod, often pay for themselves on energy savings alone, with the downtime prevention effectively free.

Sign 7: You’ve tried predictive maintenance before and it “didn’t work”

Maybe you bought sensors that generated noise instead of insight. Maybe the data went to a dashboard nobody looked at. This is common, and it puts sites off for years, which is its own hidden cost: you keep paying reactive prices because one project disappointed you.

The good news is that these failures follow predictable patterns, and they’re avoidable. We’ve written up the most common ones here: Common predictive maintenance failures (and how to avoid them). The short version: success depends less on the technology and more on choosing the right assets, the right technique, and a partner who interprets the data for you rather than handing you a firehose.

So what should you do about it?

You don’t need to rip up your maintenance strategy overnight. A sensible path looks like this:

1. Put a real number on your downtime. Use the benefits calculator or your own figures. You can’t prioritise a cost you haven’t measured.
2. Identify your critical few. Which machines would hurt most if they stopped? Start there, not everywhere.
3. Get a baseline health check. An on-site diagnostic survey will tell you the current condition of your key assets, often revealing problems you didn’t know you had.
4. Monitor what matters. Continuous monitoring for the critical machines, portable checks for the rest.
5. Act on the findings. The savings come from planned interventions, not from the data itself.

If you want a fuller picture of what the payback looks like, this piece on the 7 benefits of condition-based maintenance covers it honestly, including where it isn’t worth it.

The bottom line

If your team spends its days firefighting, your PMs run on a calendar, failures “come out of nowhere”, and nobody can tell you the health of your critical assets today, then a machine shutdown is costing you money whether it happened this week or not. You’re paying in reactive repair premiums, wasted planned maintenance, tied-up spares, higher energy bills, and the ever-present risk of the big one.

The fix isn’t more maintenance. It’s smarter maintenance, guided by the actual condition of your machines.

Want to know what your machines are trying to tell you? Faraday Predictive’s systems install easily, monitor from the switchroom without touching the machine, and come with expert interpretation, not just data. Book a chat with the team and find out what a health check of your critical assets would show.

FAQs