Free and Open Source CMMS: What You Give Up

Every few months someone asks me the same question, usually after a budget request has been declined: can we just use a free CMMS? The honest answer is yes, sometimes, and the people who say otherwise are normally selling something. What nobody tells you is that the word "free" is doing three completely different jobs in the market at once, and the decision you actually face depends entirely on which of the three you are looking at. A self-hosted open-source project, a permanently free tier of a commercial SaaS product, and a 30-day trial are not variations on a theme. They fail in different ways, they cost you in different places, and they suit different organisations.

The message up front: free CMMS software does not remove cost, it relocates it. With open source you take on hosting, backups, upgrades, security patching and a named owner. With a free tier you accept caps and missing capabilities, usually the integration, mobility and audit features that matter most later. With a trial you get everything and thirty days. For a single site with under a dozen assets-per-person and strong internal IT, that trade is often good. For anything audited, integrated or multi-site, it is usually not.

1. The three things people mean by "free"

Before you compare anything, work out which category you are in. I have watched buyers spend weeks evaluating an open-source project when what they actually wanted was a free tier, and vice versa. They are answering different questions.

• Genuine open source you self-host. The source code is published under a licence that lets you run, modify and redistribute it. There is no licence fee, ever, for any number of users. You download it, you install it on a server you control, and everything after that is yours: the database, the backups, the upgrades, the security patches, the customisation. The software is free. The operation is not.
• A free tier of a commercial SaaS product. The vendor hosts it, maintains it, patches it and backs it up. You pay nothing, but you accept limits: a user cap, an asset cap, a feature ceiling, or all three. The business model is that you outgrow it. That is not cynical, it is just how the product is designed, and it is fine as long as you know the growth path and its cost.
• A time-limited trial. Full product, full features, 14 to 30 days, then it stops or converts. This is not a way to run maintenance. It is a way to evaluate software, and it is genuinely valuable for that. Confusing a trial with a free product is the most common and most expensive of the three mistakes, because you migrate real data into something you have no licence to keep using.

There is a fourth thing that gets mislabelled as free: a spreadsheet. If you are currently running maintenance in Excel, everything in this guide applies to your decision, but be clear that a spreadsheet is not a free CMMS. It has no work order state machine, no audit trail, no scheduling engine and no multi-user integrity. For what a CMMS actually is and the modules that define it, start with the complete buyer's introduction and core modules explained.

2. What is genuinely out there in open source

The open-source maintenance and asset management space is smaller and quieter than the open-source world generally. There is no Linux of CMMS, no project with the community depth of PostgreSQL or the ubiquity of WordPress. What exists is a handful of serious projects, a long tail of abandoned ones, and a number of "open source" claims that turn out to mean "open core with the useful modules behind a paid edition".

Projects I am comfortable naming, because they have a real track record rather than a GitHub page with two years of silence:

• openMAINT , an open-source property and facility management application built on the CMDBuild framework by Tecnoteca in Italy. It is the closest thing in the open-source world to a real CAFM-leaning CMMS: asset register, maintenance scheduling, work orders, spaces, and documents. It is also the most honest illustration of the trade in this article, because it is powerful and it is configuration-heavy. You are not clicking through a wizard.
• CMDBuild , the underlying open-source asset and configuration management framework. Worth knowing about even if you choose openMAINT, because the data model flexibility, and the effort it demands, comes from here.
• Odoo , which publishes a Community edition under an open-source licence and includes a maintenance application alongside inventory, purchasing and accounting. If you already run Odoo for the business side, the maintenance module is a reasonable place to start and the integration problem largely disappears. Read the edition boundaries carefully: Community and Enterprise differ, and some capabilities people assume are included are not.
• IT-oriented asset tools such as the well-known open-source IT asset and inventory projects. I would treat these as adjacent rather than equivalent. They are strong on asset tracking and licence management and generally thin on preventive maintenance scheduling, spare parts issue, and craft labour. If your "assets" are laptops, they may fit. If they are chillers, they usually do not.

Beyond that, I would rather hedge than assert. The category has a long history of projects that looked promising, attracted a small user base, and then stalled when the maintainer's circumstances changed. If you find a project I have not named, judge it on evidence rather than on the readme: commit activity in the last six months, more than one active maintainer, a real release history with version numbers and changelogs, a forum or issue tracker where questions get answered, and documented upgrade paths between major versions. Absence of any one of those is a warning. Absence of three is a decision.

3. The true cost of self-hosting, itemised

This is the part that gets skipped. When a maintenance manager builds the business case for a free CMMS, the licence line reads zero and the rest of the page is blank. The rest of the page is where the money is. Here is what actually lands on your side of the line, and it is worth walking through this exercise properly alongside what maintenance software really costs, because the comparison only means something when both sides are costed the same way.

None of these is an argument against self-hosting. They are the specification for doing it properly. The organisations I have seen succeed with open source did not treat it as a way to avoid cost, they treated it as a decision to spend internally rather than externally, and they staffed accordingly. The ones that failed treated a zero licence fee as a zero total cost, and discovered the difference eighteen months later when an upgrade was overdue and the person who had built it had moved on.

4. What happens when that person leaves

Nearly every self-hosted CMMS I have encountered in a mid-sized organisation was set up by one capable, enthusiastic individual. A maintenance engineer who taught himself enough Linux. An IT generalist who liked the project. It worked, sometimes very well, and for two or three years it was the best system that organisation had ever had.

Then that person resigned, retired or was promoted, and the organisation discovered that the entire system existed in one person's head. Nobody else knew where it was hosted, what the admin credentials were, how the PM schedules had been configured, whether the backups ran, or which version it was on. The system kept working, because well-built systems do, until it did not, and at that point there was nobody to call.

The mitigations are unglamorous and they work: a documented runbook covering install, backup, restore and upgrade; credentials in the organisation's password manager rather than someone's notebook; a restore rehearsed to a clean server at least annually; and a second person who has actually performed that restore rather than read about it. If you are not prepared to fund those four things, a hosted product is the cheaper option regardless of what the licence line says. The cloud versus on-premise comparison works through the same trade in more depth.

5. Free tier limits that actually bite

Free tiers of commercial SaaS products solve every problem in the section above. The vendor hosts, patches, backs up and upgrades. What you give up instead is capability, and the pattern across the market is consistent enough to predict. Vendors do not put arbitrary limits on free tiers. They put limits exactly where the product becomes load-bearing for a larger organisation, because that is the point at which they want you to pay.

Several established vendors, MaintainX and UpKeep among those that have offered free plans at various points, use some combination of the following. Check the current terms directly rather than trusting any article, including this one, because tier definitions change often.

• User caps. Often two to five full users. This is the limit that bites first and hardest, because the value of a CMMS is proportional to how many technicians actually record work in it. A two-user cap means one planner and one supervisor, and the technicians stay on paper.
• Asset or work order caps. Sometimes a hard asset count, sometimes a rolling monthly work order volume. Model your real numbers, including reactive jobs, before you assume you fit.
• No API. This is the quiet one. Without an API you cannot push assets from finance, pull meter readings from a BMS, or feed a corporate dashboard. Everything becomes manual re-entry, and manual re-entry decays.
• No offline mobile. Critical in plant rooms, basements, lift shafts, tunnels and remote sites, which is to say most of the places maintenance actually happens. An online-only app in a building with no signal in the riser is an app technicians stop opening.
• No single sign-on. Free tiers almost never include SSO or directory integration. That means separate credentials, manual onboarding, and, importantly, manual offboarding, which is exactly the control an auditor will ask about.
• Restricted reporting. Usually a handful of fixed reports with no custom builder and no scheduled distribution. You can see the data on screen and you cannot get the view your regulator or your board wants.
• Thin permissions. Often a single admin role and a single user role, with no ability to separate who approves work from who performs it. Where segregation of duties matters, see role-based access control design.
• Data export limits. The one to check hardest. Ask specifically: can I export every table, including attachments and full work order history, in a structured format, at any time, without paying? If the answer is vague, treat the free tier as a place you cannot safely leave.
• No attachments or tight storage limits. Photographs and signed-off checklists are often the most valuable evidence a work order carries, and they are storage-expensive, so they get capped early.

6. The honest case for free

Having spent three sections on what free costs, let me be equally clear that there are situations where I would actively recommend it over a paid product, and where paying would be the worse decision.

• Very small teams on a single site. One building, one maintenance team of three to eight people, a few hundred assets, no external audit obligation. A well-chosen free tier will comfortably run this and will be a significant improvement on the spreadsheet it replaces. The user cap is the thing to check, nothing else.
• Proving the concept before a capital request. This is the strongest case of all. Run a free tier or a self-hosted instance on one asset group for three to six months. You come out with a real asset register, real PM compliance figures and real evidence of how much reactive work you were absorbing. That evidence is what gets a paid system approved, and it is far more persuasive than a vendor's ROI calculator. It also teaches you what you actually need, which changes your requirements list, usually by shortening it.
• Organisations with genuine internal IT capability. If you already run your own Linux servers, already have a patching regime, already test restores, and already have two people who could rebuild a web application, the marginal cost of hosting one more is small and real. Self-hosting is a rational choice here in a way it simply is not elsewhere.
• Data sovereignty or network isolation requirements. Where policy genuinely prohibits maintenance data leaving your infrastructure, or the site has no reliable internet, self-hosted open source may be the only option that complies. That is a constraint, not a preference, and it deserves to be funded as such.
• Where you need an unusual data model. Open source lets you change the schema. Almost nobody should exercise this, but the small number of organisations with genuinely non-standard asset structures sometimes find that flexibility worth all the operational overhead.

The pilot case deserves one more sentence, because it is the one I recommend most often. A free pilot that produces a clean asset hierarchy and six months of honest work order data has already done the hardest part of any implementation. Whatever you buy afterwards, you will migrate that, and you will migrate it faster and better than an organisation starting from paper. Treat the free tool as the first phase of the project rather than a substitute for it, and plan the handover from the start, using the step-by-step implementation plan and the data migration strategy.

7. Fit and no-fit, in one table

Take your own situation down the left column and be strict with yourself. Where three or more rows land in the right-hand column, the free route will cost you more than it saves.

8. How to evaluate a free option properly

The discipline is the same as for a paid purchase, and skipping it because nothing is being invoiced is how organisations end up migrating twice. I would run this sequence, and it maps onto the wider method in how to shortlist CMMS software and the scoring framework.

1. Write down the ten things you need it to do. Not fifty. Ten, in your own words, ordered. Most free options will satisfy six or seven, and which three are missing decides everything.
2. Count your real users, honestly. Include every technician who should be closing their own work orders, not just the planners. This one number eliminates most free tiers immediately, and better to know now.
3. Test the exit before the entry. Load fifty assets and twenty work orders with attachments, then export everything and inspect what you get. If the export is incomplete, stop. Nothing else about the product matters.
4. Do the mobile test in the worst place you have. Give a technician a phone in the plant room with the poorest signal on site and ask them to close a job with a photo. Their reaction is your adoption forecast.
5. For self-hosted, restore before you go live. Install, load sample data, take a backup, then restore it onto a different server. If you cannot complete that in a day, you are not ready to operate it.
6. Check the project's pulse, or the vendor's tier history. For open source: recent commits, multiple maintainers, real releases. For a free tier: has the vendor narrowed or withdrawn it before? Both tell you about the next three years.
7. Price the paid path now. Get the cost of the tier above, per user, in writing, before you commit data. You are choosing a growth path, not just a starting point.
8. Name the owner and the deputy. In writing, with the runbook location. If the role is vacant, the project is not ready to start.

For the specific question of which products suit a small maintenance team, paid and free, the detail sits in the best CMMS for small teams.

9. The security conversation you will eventually have

This section exists because it is the one most often skipped, and the one with the least forgiving consequences. A self-hosted CMMS is a web application holding your asset register, your building layouts, your contractor details and often your access procedures. If it is reachable from the internet so technicians can use it from site, it will be scanned, automatically and continuously, by parties who have no idea what it is.

The controls are not exotic, but they must be somebody's actual job: patch the application and its framework on the vendor's cadence rather than annually; keep the database off the public internet; enforce individual accounts with multi-factor authentication for anything internet-facing; remove leavers promptly, which is much harder without directory integration; encrypt backups and store one copy off the host; and keep an access log you could hand to an auditor. The OWASP Foundation publishes the standard, freely available references for what web application security requires, and it is the right baseline for a self-hosted deployment.

This is the honest asymmetry between the two free routes. A reputable SaaS vendor does all of the above as a condition of staying in business, and their free tier inherits it. When you self-host, you inherit the obligation instead. That is entirely doable, and it is not free.

The idea to walk away with

Free CMMS software is not a lesser version of paid CMMS software. It is the same category of product with the operating cost moved from a vendor invoice to your own organisation, or with capability removed until you pay. Both are legitimate trades, and both are good decisions in the right circumstances: a single site with a small team, a pilot that will build the business case, an organisation with real internal IT capacity, or a genuine data sovereignty constraint.

What makes free go wrong is never the licence. It is treating a zero licence fee as a zero total cost, and then discovering the bill later in the form of an unpatched server, an untested backup, an audit you cannot evidence, a user cap that keeps technicians on paper, or an export you cannot get. Cost the whole thing, name the owner and the deputy, test the exit before the entry, and free becomes a sound engineering choice rather than a deferred problem.

Final thoughts

If I had to compress the advice to one recommendation: use free deliberately and for a defined purpose. Use a free tier to run a genuinely small operation well, or use either route as a three-to-six month pilot whose explicit output is a clean asset hierarchy, honest work order data, and a shortened requirements list. Both are excellent uses of free software. Drifting into a free tool as a permanent solution for an organisation that has outgrown it is not, and the cost of that drift shows up as data you cannot trust, which is the one thing no CMMS can fix later.

The organisations I have seen do this best were not the ones with the biggest budgets. They were the ones who were clear-eyed about which of the three "frees" they were using, what it did not cover, and who was accountable for the gap.

On independence: this is a practitioner's assessment, not a paid review. No vendor or project named here has had editorial input or a commercial relationship with this publication, and nothing here is sponsored. Tier definitions, licences and project activity change, so verify current terms directly with the vendor or project before deciding.

Related reading: What is a CMMS: a complete buyer's introduction, Best CMMS for small teams, CMMS pricing: what it really costs, Cloud-based CMMS vs on-premise, How to shortlist CMMS software, A CMMS scoring framework.