Recent research shows that 51% of IT budgets will go to cloud-based systems development and operations by 2025. No surprises there; we all know this day is coming. We also know that the cloud budget percentage will continue to increase long after 2025. Is it time to change how we think about cloud computing?
Enterprise IT spending on public cloud computing will overtake spending on traditional IT in 2025, within addressable segments, according to Gartner. The research includes only those enterprise IT categories that can transition to cloud: application software, infrastructure software, business process services, and system infrastructure markets. Gartner found that by 2025, 51% of IT spending in these four categories will have shifted from traditional solutions to public clouds. This is compared to 41% in 2022.
There’s no real news here. I’m sure cloud computing market watchers understood that we would hit the tipping point at some time. However, I’m not sure that they understand what this should mean to enterprise IT. It’s something we should discuss.
Keep in mind that these numbers focus on the transition to public clouds not private ones. They concern the migrations from traditional platforms to public cloud providers and do not include those who just install software in a data center and call it cloud.
This is a big problem when you look for patterns around cloud adoption. Most technology vendors cloud-washed their stuff so it would be counted in the public cloud providers’ market. This gives us overstated cloud market figures.
Nevertheless, spending more on public cloud services than traditional IT requires some rethinking that needs to happen between now and 2025. I suspect that if asked about the future use of public cloud providers, most CIOs would say that it will be less than traditional IT spending in 2025. In other words, they may not yet see this coming.
Are CIOs missing today’s clear trends? Cloud can slowly take over, and companies sometimes have little internal understanding of how deep the entire enterprise is into public clouds. Most CIOs do not understand how much public cloud use exists within their own companies until they do surveys or scan the network. It’s just too easy for anyone in a company to become a public cloud computing customer, in many cases, without IT leadership even knowing. We call this “shadow IT.” Today it’s much more common than most can comprehend.
So, if we’re going to spend more on the public clouds within three years, what does that mean for IT planning now? I have a couple of suggestions:
First, most public cloud use will expand within multicloud deployments. IT must learn how to battle overwhelming complexity and come up with a plan.
Focus on lax policies around what and how many public cloud services can be leveraged. Remember, when someone raises the battle flag for “best of breed,” the trade-off is complexity. Also, remember that more modern tools can hide complexity by using abstraction and automation. AIops tools come to mind, but there are many others.
Second, think about cross-cloud technologies and planning.
If you run 5,000 cloud services that span three major public cloud providers, find common technology to manage everything that sits above the clouds. This includes deploying operations, security, and governance as common services that run between and over all deployed public clouds.
Let’s take security. If you plan to deploy three or more security layers, such as identity access management (IAM) with native security systems running on each cloud, the resulting operational challenge will most likely lead to a breach. If you deploy more than a single public cloud, security, governance, monitoring, and operations need to be a single layer of technologies that spans all public clouds.
Of course, planning common layers that run over but not within public cloud providers requires much more sophistication. This type of planning is one of my specialties. It’s a long and detailed process to figure out your approach to public cloud management and how to deal with core services that need to span clouds.
Many people think they can look at what their peers or competitors are doing in similar-sized enterprises or even in similar marketplaces and just replicate it. Those people are way off base. “Umbrella” solutions are unique to your enterprise’s needs. Yes, I’m also saying it depends.
There are about a hundred more things enterprises need to do before they reach the public cloud tipping point, obviously much more than I could list here. These two things are a great place to start.