Cloud repatriation is real, but not real enough to change the prevailing cloud trajectory. Think of it as the exception, and not the rule.
It’s a question I’ve heard multiple times: “We heard that [insert giant company name] is taking their apps/data off [insert giant public cloud vendor name] and moving it back into their own data center. Is this the beginning of a big shift?” If your job is in any way related to selling products or services for enterprise data centers, “cloud repatriation” sounds like a promising concept. Amazon Web Services (AWS) and Microsoft have been eating the lunch of a whole bunch of IT companies, and those IT companies would like that lunch back, thank you very much. But is the exodus of customers from public cloud really happening? Well, I have some good news and some bad news.
Bad news first: Cloud repatriation is not a market-changer
Cloud repatriation is not real in the sense of being a major, market-shifting trend worthy of its own buzzword. I will not deny the existence of one-off customers making a monumental shift away from public cloud. TBR sees anecdotal evidence of companies leaving public cloud environments, but we don’t see a wholesale move to strictly on-premises environments. The numbers tell it all: TBR estimates the PaaS and IaaS market grew 16% overall in the second quarter, with the big three juggernauts (AWS, Microsoft, Google) growing 58% on average within the segment, accounting for about $10 billion of the quarterly segment revenue. If anything, the public cloud market is moving toward an oligopoly as it consolidates. But it’s not shrinking. The market growth is far outpacing the loss of any customers that may be defecting.
The good news: Companies continue to use on-premises data centers, negating the need for repatriation
Very few companies see a future without owning some kind of data center. Apps that never leave the data center do not need to be repatriated in the first place (although they will likely need to evolve to a more agile and scalable delivery method). As you can see in Figure 1, the bulk of companies expect to maintain a roughly even mix of on-premises apps and those in hosted cloud environments. Smaller companies are most aggressive in their desire to reduce their on-premises footprint while the largest companies make it clear they don’t see a future in hosting everything. These projections make sense to me, especially based on my conversations with IT execs in small and large enterprises. Smaller companies tend to be concerned about the proficiency of their own data center while larger companies are full of complexities that make moving to a new environment a challenge.
The reality: Most companies seek a balance
By and large, companies are evaluating the best fit for workloads, acknowledging that there is no one-size-fits-all solution. Regardless of the type of cloud(s) being used, more than 80% of users will either maintain or expand their environments over the next three years. The proportion of buyers planning for public cloud expansion exceeds that of those engaging in on-premises private cloud expansion. But the fact remains that there is not a mass exodus from any specific environment. Regardless of environment, changes and evolutions will occur, even within self-built private clouds.
Given that business-to-business buyers are all over the map when it comes to cloud adoption, where can IT vendors succeed? There’s no easy answer, however, when discussing this topic with my colleague Senior Analyst Cassandra Mooshian, she had this to say:
“Recognizing there will be both exceptions and changes for most customers over the next three years is important for vendors, regardless of their cloud point of view. Yes, there will be workloads that have migrated to cloud that will move back to a traditional or on-premises delivery method. However, there will also be services deployed on premises that could eventually be moved to a cloud environment as customer needs and costs change. Something simple, yet critical, for vendors is to understand that no two IT environments are the same, especially across market tiers. Vendors may want their customers to go all-in on cloud, but that just is not feasible for larger organizations or even smaller companies in regulated industries or regions.
“The key to vendor success is to understand that there will be workloads best suited for cloud, while others may work just fine in legacy environments. The kicker will be in helping customers embrace hybrid, understand what works best where, and ultimately integrate and orchestrate it across each customer’s unique blend of legacy and cloud workloads. Once trust is established and there’s a mutual understanding around the idea that all options can and should be considered, that’s when long-term relationships start, and each company has a ‘favorite’ vendor or two.”