Some statistics are cause for worry, and some statistics are truly alarming. It’s reasonable to assume that one statistic is, to the Google management, clearly in the alarming category.
That statistic is the market share held by Google Cloud Platform. GCP’s real number is debatable – the cloud market is measured many ways – but for simplicity let’s use Synergy Research’s:
If you’re a GCP executive, the numbers reveal that something very fundamental has not succeeded. Running in a distant fourth place? Ouch.
Given Google’s gargantuan strengths, what happened? Being competitive in the cloud relies on having multiple far-flung sophisticated datacenters tied together with an advanced network – which Google certainly has. And the brilliance of the Google engineers and developers is second to none, advancing science-fiction tech like self-driving cars, VR, and AlphaGo, the AI program that bested top human champions in the intuitive game of Go.
Yet as impressive as its tech prowess is, GCP’s ability to cater to the prosaic needs of enterprise cloud customers has been limited, even fumbling. Google has always focused more on selling its own services rather than hosting legacy applications, but these legacy apps are the engine that drives business. Remarkably, GCP customers don’t get support for Oracle software, as they do on Amazon Web Services. Alas, catering to the needs of enterprise clients isn’t about deep genius – it’s about working with others. GCP has been like the high school student with straight A’s and perfect SAT scores that somehow doesn’t have too many friends.
As cloud computing continues to rocket forward, Google’s hyper-focus on its own brilliant tech is costly. Gartner forecasts that the public cloud services market will grow to $204 billion this year, up 16.5 from $175 billion last year. Upping the stakes, dominance in the cloud translates to revenue in related areas like Big Data and Artificial Intelligence.
Given GCP’s struggles, a core question seemed to hang over the recent GCPNext, Google’s annual cloud showcase: will Google Cloud Platform succeed in the enterprise?
It was clear that Google executives walked on stage with a unified message to address that issue. “We’ve come a long way as an enterprise company,” said Google CEO Sundar Pichai, in a statement that seemed as aspirational as currently accurate. Most direct was Eric Schmidt, executive chairman of Google parent company Alphabet. He was frank about the limits of App Engine, an earlier cloud effort. In this new era, “We’ve decided to meet you where you are, as opposed to where we think you should be.”
Accompanying the executives’ pitches were demonstrations of Google’s next-gen web-scale applications for machine learning and deep analytics. Also highlighted were cloud-focused advances based on Google’s search business, like image recognition and processing. To spotlight its enterprise focus, the event offered splashy videos from GCP clients Spotify, Disney Interactive and Coca-Cola.
Afterward I spoke with Diane Greene, SVP of Google’s cloud business. Greene, co-founder and former CEO of VMware, was hired to run GCP in a move that signaled its sharp focus on business clients. In Silicon Valley, Greene is rightfully known as a legend. I asked her about changes in GCP since her tenure began last November.
“The engineering is ramping [up]…but we’re paying a lot more attention to how to interact with our customers and partners,” she told me. This isn’t limited to large enterprise – small and medium sized business are also targeted, she said. As for supporting software like Oracle’s, geared for traditional business clients, she said, “We’ve got to build out our ecosystem – we’re very focused on it.”
Executives present at GCPNext16
Despite these assurances, the event hosted plenty of experts who voiced skepticism about Google’s enterprise prospects.
“They must do a much, much better job of appealing to the enterprise,” said Holger Mueller, Constellation Research analyst. “These great wins they have, Spotify, Coca-Cola, and Disney Interactive? The average company cannot relate to the loads and profiles they have. You have Google saying they’re typical problems, but they’re not typical problems, at least in the minds of the CIOS and CTOs listening.
“To me, it always looks like the Google people are super smart, super intelligent, and they think the enterprise will figure it out for themselves,” Mueller said. Yet enterprise clients are typically overwhelmed – with a busy workload and new technology. Google “has to bridge that,” he said,
John R. Rymer, Forrester analyst, questioned Google’s approach to the enterprise. He echoed Mueller’s sentiments that Spotify and Disney Interactive don’t represent the typical corporate workload. “Is the ambition to go after much more prosaic applications involving relational databases and transactions and integration? Or is it really going after those super-scale digital projects, to really try and win there? It’s hard to win everywhere.”
The contrast between Google and AWS is pronounced, Rymer said. “We went up to AWS re:Invent, talking to a whole bunch of customers up there. They’re excited about moving their Oracle databases into MySQL on AWS. It’s not the [Coca-Cola] Happiness Project.”
However, he sees Google’s progress toward business customers. “They’re starting to break out all these services, which previously were all kind of buried. So they have a big service catalog – gee, it looks like AWS. Used to be, there were a small number of services. And they’re really starting to refactor and bring out more APIs and more services. Looks a lot like the competition looks.”
GCP executives, said Sam Charrington, CloudPulse Strategies consultant, seem to view cloud as yet another markets that Google has entered late, but can still win based on product strength. Google’s success with search, mail, maps, browser and mobile do support this idea. But, he noted, these are all consumer markets. “No doubt this strategy has yielded some innovative offerings – particularly in the data and analytics space – yet their ability to win enterprise markets remains largely unproven.”
A user tries Google Cardboard, the company's low-cost virtual reality device.
Perhaps GCP won’t need to so strenuously court enterprise clients to win cloud share. Al Hilwa, IDC analyst, speculated that Google, “may be thinking, they’re all going to come our way.” GCP, in other words, may draw a large corporate user base due to the sheer innovation of its technology. In the current climate, exceptionally advanced technology isn’t only for startups.
“Enterprises are trying their best everyday to look like successful startups,” he said. “All these companies are setting up innovation teams, transformation teams. Companies like Walmart and Target are setting up shops in Seattle, the Bay Area. Every category has a disruptor, therefore all the established players are scrutinizing their disruptors and trying to imitate.”
Even established legacy players realize they need to look toward the edge. To be sure, Hilwa said, AWS and Microsoft Azure use a more enterprise-friendly focus, and in reality that’s needed to succeed.
However, the importance of leveraging the most advanced tech will grow far greater in the years ahead. This favors Google.
If I had to predict which cloud platform will dominate several years down the road – a perilous forecast, given the variables – I’d say Google has a superb chance. Certainly that’s a contrarian position, in light of GCP’s struggles.
GCP’s long term edge will rely on two factors: 1) its new management, with its willingness to admit a course correction, and 2) most important – the quality of its technology.
For the management piece, hiring Diane Greene to run GCP will likely pay enormous dividends. Her name isn’t as marquee as other tech luminaries, but she has done as much as anyone to create the modern IT world. As co-founder and CEO of VMware she ushered in the era of virtualization – which enabled cloud computing. And she knows enterprise buyers; VMware is the archetypical enterprise company. I heard at CCPNext that she has bulked up the sales staff and is rallying them to more aggressively court business customers.
As for GCP’s edge in technology, it’s clear: the company has always been profoundly focused on the tech itself. And the benefit of that was clear as I watched the morning demo at CCP. I’ve sat through countless tech demos – enough to be jaded – and I’ve never seen one at that level. GCP’s advances in machine learning and analytics are truly at the edge.
Of course Google isn’t alone in its pursuit of machine learning or analytics; Microsoft, AWS, and IBM all have impressive efforts. But Google, with its legacy strength in the network of the Internet and its focus on futuristic tech, is arguably better positioned than any of its competitors, long term.
GCP is late but has finally very much woken up. As the years roll forward and leveraging advanced tech becomes a competitive necessity for the enterprise, Google Cloud will be a force to be reckoned with.