Some of these might seem obvious, and people sometimes play them lip service but never really go far enough under the covers. One is the architecture. You'll hear a lot of vendors throw large numbers at you. I can monitor x hundred thousand elements or x hundred thousand devices, which is great. A lot of people go check the box and move on. However, it's really important to understand the architecture they're using in order to actually achieve that. Because otherwise, when you get to the sort of scales that some of the large organizations are, you are going to have bottlenecks. When you've made that sort of investment in the product, you need the source of truth available and the data available at your fingertips all the time.
The historical tiered architecture of these type of PM solutions and even FM solutions hasn't changed a lot over the years. You have collectors that will go out to an aggregater that will go to a reporter. That's a lot of tiers to get the data from. You need something that's going to be a little more cutting edge, something that's a little different, that's going to allow you to scale to a global reach and be sure that you're going to get the data you need when you need it.
The other thing people often forget is visualization. There's products out there, I know, that will guarantee and collect every single point of data that you want but then cannot present you that data in any consumable or meanable format. That's no good. Similarly, there are other products out there that will give you great visualization tools but do not collect all the data. Both of those approaches are equally useless. You need something that you know is going to collect the data all the time but has resilience and that you're going to be able to have consumable data that you can easily visualize for operational benefit.