Choosing a DCIM – common pitfalls and problems

27 February 2025

Martin Docherty, Director of Inside Sales, Assetspire Limited

Martin Docherty, Director of Inside Sales, Assetspire Limited

Not all DCIMs are created equally. Some are a patchwork (or bucket full) of many other products, some are more like a single platform. Some vendors charge per asset or rack or monitoring point, some charge per user.

For the purposes of this article, let's agree that DCIM software should be delivering accurate, up to date and actionable data from your data centres and digital infrastructure, from M&E, through to IT, network and building fabric. A DCIM platform should be providing you with data you can trust and a view across your whole estate, site by site, incomer to rack, floor by floor, room by room, CRAC by CRAC etc.

Every company has slightly different requirements, challenges and problems, and it's impossible to cover them all here, so instead, I want to focus on where the common complaints and pitfalls are with DCIM Software, regardless of the customer specific requirements and challenges.

The most common complaints we hear from DCIM users are around the usability and the ability to collect and maintain accurate data.

On the data quality side, if you lose trust in the data, then there’s little point in having it. Decisions made with poor data are simply poor decisions - you just don’t know that, yet. Check what tools are included to ensure data remains accurate. It’s not just about taking a Modbus or BACnet data feed for power consumption, it's also about knowing where everything is, how your assets are being used, how they are connected to each other, criticality rating, age, condition, servicing and maintenance schedules, cost to replace/maintain/repair, MOPs/SOPs etc. What happens when a filter gets changed on a CRAC? What happens when a few servers come out of the lab and into production within a rack? Or a cross-connect/meet-me room connection changes, how are you capturing this accurately and completely? Risk and costs can silently climb up if you’re not tracking things properly, and we’ve seen many situations where the cost increases due to poor data dwarf the cost of the DCIM software.

A single source of truth is necessary for a data centre estate to be run efficiently and for teams to be able to collaborate effectively, and the DCIM should offer this. Having multiple versions of the truth across several departments will increase risk, costs and complexity and kybosh your ability to make good decisions in good time. We’ve seen some of the biggest colocation providers struggle with financial planning and refreshes, because the asset data just isn’t up to scratch. If your data is good, then you can plan with confidence, and spot opportunities to remove risk, costs, and complexities and you can deliver a better service to your customers.

On the usability front, if you’re constantly having to train/retrain staff on how to use the software, you’re going to struggle. It should be intuitive, and not overly complicated to use. We’ve seen DCIM platforms thrown out for this very reason. There’s a shortage of talent across the industry right now, and these tools should be allowing staff to do more, not taking up their valuable time with badly designed, clunky software.

Another (often overlooked when purchasing) issue we see in the market is the significant amount of time it takes to deploy a DCIM platform. Anything upwards of 18 months is not uncommon, and this is often frustrating and costly for the customer, and of course, risky. Choosing a platform that is quick to deploy and starts delivering value and insights is key.

Trying to understand what your emerging challenges and complexities will be is also important when considering investing in a DCIM tool. As an example, in the last year or two, we’ve seen colocation tenants demanding accurate and granular data for their own reporting (e.g. Scope 3). The days of just billing for power and maintaining an SLA on ambient temperatures are over. Running these reports is often time consuming. Another example is with new operators coming into the space, building out GPU-as-a-Service infrastructure at breakneck speed, which brings its own unique challenges (e.g. more complicated supply chain, or complexity of connectivity chains).

These are all problems that can and should be solved with a proper DCIM platform and tools.