How to calculate the return on a CMMS

When you go looking for the return on a maintenance system, savings percentages show up everywhere. We don’t publish any, and it’s worth explaining why before proposing an alternative.

That figure doesn’t exist. It depends on the sector, the starting point, the size of the fleet, and above all on whether the team actually uses the system. A percentage pulled from someone else’s study says nothing about your case, and citing it is the fastest way to lose credibility when someone asks you to justify it six months later.

What you can do is a calculation with your own data. It’s more work and it’s worth infinitely more.

What you’re already paying for

Four line items, all in last year’s invoices.

The urgency surcharge. The same intervention costs less when scheduled than when it’s an emergency: no immediate dispatch, no rush shipping for parts, no forced shutdown, and no other job left undone that day. Check your unscheduled-intervention invoices.

Downtime. It usually outweighs the repair cost and almost never gets charged to maintenance. Estimate the cost of an hour of downtime in your operation.

Second visits. Going back because of missing material or missing information gets paid for twice. Ask your technicians how many happen per month; they know exactly.

Warranties paid out of pocket. Take a batch of repair invoices and compare them against the equipment’s purchase date. It shows up with uncomfortable frequency.

What you save on administration

Here there are hours that disappear entirely, not just hours that go faster:

Transcribing job reports. Chasing receipts. Answering calls that are only asking how something’s going. Retyping into the ERP what was already written down that morning.

Estimate how many hours a week currently go into that. It’s a figure that surprises people once it’s actually counted.

What it costs

For the calculation to be honest, the other column needs to be complete too:

The tool. And here the pricing model matters: if it’s per user, do the math with your headcount three years from now, including customers and subcontractors. In GMAO Cloud licenses are unlimited across all three plans.

Infrastructure, if the system is on-premise: server, backups, updates and whoever maintains it. In the cloud, that line item doesn’t exist.

Implementation. The largest item, and the one that’s always underestimated: setting up assets, defining checklists and periodicities, and getting the team used to recording in the field. It exists with any tool.

The team’s time during the first few weeks.

What you can’t put in the calculation

Worth saying, because it’s what gets inflated most in other people’s figures.

Equipment lifespan. It’s real — wear addressed in time doesn’t turn into a breakdown — and it can’t be quantified beforehand. What you can do is start measuring it: the accumulated cost per asset in the reports gives you the series that, in two years, will let you compare.

The reduction in breakdowns. It depends on the plan targeting the right failure points, and that gets adjusted with history you don’t have yet.

The value that isn’t financial

Two things that don’t fit in a spreadsheet and that, in many companies, weigh more than the savings.

Being able to prove it. If there’s equipment subject to regulation, or customers demanding proof, the system stops being a matter of efficiency and becomes one of risk. Legally required maintenance — which isn’t a module, but the preventive maintenance mechanism paired with the checklist the regulation requires — rests on the history of work orders, checklists and documentation with its dates.

No longer depending on one person. When what needs to be checked on each piece of equipment is written down by family, the knowledge doesn’t leave with whoever quits.

The cost of waiting

The argument that carries the most weight and gets used the least: the history you’re not building now can’t be recovered later.

Today’s orders can still be recorded tomorrow with some effort; what happened last year, if it wasn’t written down, doesn’t exist. Every month that passes is a month of data you’ll never have to decide whether to repair or replace a piece of equipment.

When the math doesn’t work out

For the sake of honesty, because there are such cases.

With few pieces of equipment, one site and one person handling everything. A well-maintained spreadsheet can work for years. Its problem isn’t that it’s a bad tool: it’s that it has a ceiling, and it doesn’t warn you when you reach it.

If everything is outsourced and your role is just receiving invoices. What you need first isn’t a system: it’s requiring your provider to hand over job reports for what they do. Though that’s usually the first step toward wanting one.

If the process isn’t agreed on. If nobody today decides who handles what and at what priority, the system will record the same disorder with more precision. That conversation comes first, and it doesn’t cost money.

If nobody will be able to dedicate time in the first few weeks. Without someone in-house owning the project, implementation drags on and the return moves further away.

In any of those four cases, the honest answer is to wait.

How to check it with little risk

You don’t have to decide blind. A bounded trial — a group of critical assets and a handful of technicians, over a few weeks — answers the two questions that matter: whether the team actually uses it and whether data shows up that didn’t exist before.

That second point is the definitive argument: showing a figure the company didn’t have.

The two factors that move the outcome most

Of everything above, two things explain most of the difference between an implementation that goes very well and one that goes just okay.

Whether the field team actually uses it. It’s the dominant factor and the least evaluated when comparing systems. If recording gets pushed to the end of the day, the data exists but it isn’t accurate, and none of the figures in this article can be calculated.

Whether the initial scope is reasonable. Starting with critical assets instead of the full inventory is the difference between seeing results in weeks or exhausting the team over months.

Both depend more on how the project is framed than on the tool, and both can be checked before signing anything: bringing a technician to the demo and agreeing on a bounded starting scope.

What we can promise

That the four line items from the start of this article will become measurable, which is the requirement for reducing them. Not a percentage.

And a recommendation on how to present it internally: bring your four figures, not a brochure’s. A calculation built on your own invoices is much harder to argue with than a percentage nobody knows the origin of.

If you’d like to run the numbers on your case, you can write to us or request a demo.