I've worked with a lot of different CAFM systems over the years, across a lot of different businesses, and it's given me a genuine appreciation for how differently these platforms are built and how differently they end up serving the teams using them. There's no single "best" system. There's only the right fit for what a particular team actually needs, and that's worth understanding properly before you commit to one.
Broadly, CAFM systems sit somewhere on a spectrum. At one end, there are streamlined platforms designed to get a team up and running fast - intuitive, easy to onboard, minimal training required. That's genuinely valuable, especially for a small or growing team who need something working in weeks, not months. The trade-off is usually depth: some of the more streamlined platforms can be limited when it comes to detailed reporting, or managing SLAs and KPIs with real precision, simply because that flexibility adds complexity they've deliberately designed out. It's a common enough problem that it shows up in the research too, in one industry survey of software buyers, limited functionality and inefficiency were the two most-cited reasons people ended up switching systems, at 39% and 37% respectively.
At the other end, there are highly configurable, bespoke systems built to handle almost anything you throw at them, with complex reporting, industry specific workflows, deep customisation. The trade-off there is usually time and cost: mobilising a heavily bespoke system properly can take months (if not longer!), and the platform is only as good as the configuration work that goes into it. Get that part right and it's a genuinely powerful tool. Rush it, and you can end up with a very expensive version of the same limitations you were trying to avoid.
Then there's the layer everyone's talking about right now: AI. Most CAFM providers are building in AI-supported features - predictive maintenance flagging, smarter scheduling, automated summaries. Some of this is genuinely useful and will only get better. But it's worth evaluating on its own terms, separately from the core question of whether the system reports and manages the basics the way your team needs. A helpful AI feature sitting on top of the right system is a real advantage. The same feature sitting on top of the wrong system is still worth having, but it isn't a substitute for getting the fundamentals right first.
Worth watching separately from all of this, though (and something im super excited about!), is a newer category: systems being built from the ground up with AI at their core, rather than AI bolted onto a platform designed years before it existed. That's a genuinely different proposition - the workflows, the reporting, the way information moves through the system are all shaped around what's possible with AI from day one, instead of retrofitted around it after the fact. It's early days, and it won't be the right fit for everyone just yet, but it's one of the more exciting shifts happening in this space right now, and one worth keeping an eye on even if you're not ready to make that jump today.
So how do you actually work out what's right for you? A few questions tend to cut through the noise: What do you need to report on regularly, and to whom? Who's actually using the system day to day, what do they need from the system, and how much time do they realistically have to learn it? How quickly do you need to be live and adding value, versus how much configuration are you genuinely prepared to invest in upfront? And which parts of your operation are truly industry-specific, versus just "how we've always done it"?
What I hear consistently from clients is that they're looking for a particular balance: simplicity that means their team will actually use the system without a lengthy training process, combined with just enough flexibility to reflect their industry and their specific compliance requirements. Not fully bespoke, and not a rigid, one-size-fits-all template either, something in between, chosen deliberately rather than landed on by accident.
That's really the point of a proper systems review: not to push you towards a particular platform, but to work out what you genuinely need first, so that whichever system you end up with - the one you already have, or a new one - is chosen against your actual requirements rather than a demo that looked impressive on the day.