Why you shouldn't fear a "sink or swim" cloud adoption (it beats these four alternatives)

In “Life of Pi,” the award-winning book-turned-movie, the title character, a 16-year-old shiphand named Pi Patel, finds himself adrift at sea on a raft of flotsam beside a proper lifeboat. He cannot board the lifeboat, at least not initially, because it has another passenger already: a fully grown Bengal tiger.

Pi must decide if, and how, he can survive alongside this menacing creature and the equally menacing ocean. It’s the classic “Type 1” decision that’s irreversible — get in the boat, or go down with the debris. Contrast that with a “Type 2” decision, which you can reverse later if needed. 

While rarely as dire as Pi’s choice, similar decisions take place almost every day in the business world, including when it comes to cloud adoption (migration is but “jumping ship” by another name). Recently, I’ve spoken with some companies going through their own evaluations between a sink-or-swim approach versus hedging their bets and leaving an escape route.

In one customer roundtable, I heard a few cloud customers talk about their digital transformation journey. A leader from one large company described taking an aggressive approach and told their teams that the data centers were shutting down on a certain date, and everything needed to be in the cloud. Sink the data center, so to speak, and provide only one way forward. 

I like it. We’ve seen a handful of Google Cloud customers go all-in on cloud, and when they do, it seems to reduce uncertainty and provide clarity to teams, so everyone can work towards the same objective.

But frankly, most enterprises I’ve observed take a different approach. They’re using Type-1 decision complexity — methodical, careful planning — but hoping for a Type-2 outcome. Their cloud journey is “hybrid” or “multicloud,” and like another company at that aforementioned customer forum, they believe they’ll “always have workloads on-premises.” These companies hedge their bets, and many of the leaders seem to want the freedom to reverse course if they deem it necessary.

Yet just like Pi's paradox, turning back is rarely as easy or straightforward as it seems. Forging ahead isn’t the only way to go, but it’s often the best.

The five paths to cloud adoption

So besides “sink the ships”, there are four other adoption patterns I’ve given names to. 

First, there’s “cap and grow.” The company freezes onsite investments, and puts new workloads in the cloud. It’s tough to sustain this unless you also have a plan to get those on-premises workloads migrated eventually.

The next pattern is “extend the foundation.” Here, the on-premises environment is the corporate anchor, and the tools and platforms add “the cloud” as another target. You often see a lowest-common-denominator approach as the company treats cloud providers like data center extensions. A notable problem is the skills gap is exacerbated. 

Another pattern is “parallel paths.” The company has different parts of the org doing different things. On-premises is growing, but so is consumption on one or more clouds. Eventually this comes to a head and a new leader comes in and tries to get a handle on it. 

“Selective services” is another pattern. In this case, the customer uses unique cloud services (e.g. BigQuery, AI/ML) but still keeps their “important” stuff on-premises.

Marching into the multicloud future

If I asked you to self-select into one of these journeys, how would you answer? 

Some folks have told me “all of the above,” which isn’t really a choice at all. You’ve got a big decision, and an important one. Each of these journeys involves unique hiring, training, product selection, and architecture strategy. 

Mark Twain (or was it Andrew Carnegie) once wrote, “Put all of your eggs in one basket, and WATCH THAT BASKET.” More than a century later, the advice still stands. Hedging is expensive and spreads your team thin. 

While it can seem scary, having the confidence to sink your ships and march in unison towards the cloud can be a powerful, empowering decision — I’ve seen it first hand, time and again, from leading organizations now at the forefront of innovation in their industries. 

Go all in. Focus your team, simplify your architecture, and don't waste time building escape hatches when you should be busy building your future.

For more of insights on the hybrid cloud journey, please join Google Cloud leaders and experts for the #MulticloudMindset series on Twitter Spaces, where you can catch in-depth discussions of the latest trends and even share your own questions on where the cloud is going next. And once you know what you're looking for, you can try out one of Google Cloud's many multicloud solutions for your organization.