Google I/O can be mystifying to someone with an enterprise IT perspective. Google's two-hour keynote to its annual company developer outreach (and technology roadmap) event featured a new connected-home virtual assistant (think Amazon Echo), engaging new chat software and next-generation VR technology that should replace the one that's made of cardboard.
Google has known for some time that the enterprise is where the money is. But the company has been institutionally ill-prepared to make that shift from consumer products (most of which are free, but support their advertising mission) toward enterprise-class products and customers. This is a company that sometimes develops technology for technology's sake, then devises marketable products as an afterthought. While this is going to date me, the academic aspect to Google's DNA is reminiscent of another large tech company that we've all but forgotten: Digital. To be sure, the two company cultures were/are very different. But the devotion to technology and innovation, sometimes to the detriment of market success, are quite similar.
Google has a habit of lavishing time on free products for consumers, then releasing an enterprise version with little fanfare or outreach -- then just dusting off its hands. That was the case in early 2007 with Google Apps Premier Edition, the first paid version of Google Apps and the first version aimed at business. Then as now, Google Apps for work environments costs $50 a seat per year.
I visited the Googleplex in 2007 in search of an enterprise market strategy that would support enterprise IT in the way it needs to be supported. What I found was a lot of Google naïveté about what enterprises need from enterprise apps vendors. Google Apps may be a fairly simple set of products, but that doesn't mean that enterprise IT's needs are simple. I didn't find what I was looking for.
That dusting-off-the-hands bit can be a real problem for enterprise customers. One CIO, who recently switched from Google Apps to Office365, said: "Google is trying, but they don't understand enterprise rules and policies." Plus, "we can't get our Google rep to respond in a timely manner. Sometimes he takes two to three weeks to answer and then he isn't really answering, he's just responding."
This is a recurring refrain.
Google may finally be growing up, though. In recent years it started a slow, lumbering turn toward the enterprise cloud -- but it's extremely late to the party. If it can just get its mind around the customer base, and the products and services those customers need.... Enter Diane Greene, a cofounder of VMware tapped late last year by Google to run its enterprise cloud business. Her arrival signaled a reorganization that brought enterprise apps, including Google Apps, under her direction, too. So the products are finally aligned for enterprise buyers. (And it appears that Google's mobile productivity apps are already outpacing Microsoft's.)
Yesterday, a colleague and I piled into a bus at Google I/O and were ferried with others to the Googleplex, where Diane Greene and five members of her team awaited to tell us about Google's goals for enterprise cloud, its progress, Google Apps, the TensorFlow hardware supporting its cloud effort, and in particular TPU ASIC (designed to support voice and speech recognition, text recognition and image matching) -- and anything else that we could think of.
A few things became apparent:
- Google is finally dead serious about delivering a full suite of enterprise cloud solutions.
- It has a set of smart execs in place who understand the technology and the needs of enterprise customers.
- Because it supports the cloud effort and will likely be sold to some enterprises as part of cloud solutions, Google Apps for Work is also being given direct emphasis with productivity and functionality enhancement and refinement.
The jury is still out on whether Google can figure out how to support the enterprise in two to three hours instead of two to three weeks. That's something Google is just going to have to invest in. What it has doesn't scale.
Google should also revisit issues like software licensing (Microsoft is offering valuable discounts on things like data center servers), concurrent usage limits, storage and so forth. Because many of the Google enterprise apps were built as consumer products first, they're light on management tools and administrative features, a Microsoft strength.
Finally, Google needs to have a better handle on customer pain points. For example, it rolled out the Postini anti-spam server tool under the Google label and unilaterally required all users to have a Google account to sign in to Postini. That wreaked havoc at companies where either they didn't have Gmail or not all employees had Gmail. Telling employees to use their personal Gmail accounts is not a viable solution for most CIOs.
So Google has a long upward climb ahead. It's not the No. 1 or No. 2 cloud company. It needs to develop an enterprise support system, probably with clean-sheet thinking. But it finally seems to have religion on developing and selling products to the enterprise. A sleeping giant awakes -- and hopefully that will mean more choice for IT.