Oracle OpenWorld 2016 will go down as the year CTO Larry Ellison threw down the proverbial gauntlet at Amazon's feet, announcing that Oracle is going after the cloud infrastructure market leader. History will decide whether that challenge was more bravado than reality.
As night follows day, long time Oracle critics took the opportunity to lampoon Ellison's unashamed Amazon bashing keynote while others gave Oracle the thumbs down for an unfortunate live poll that backfired on co-CEO Mark Hurd.
For my part, I saw the second Ellison keynote as classic Oracle chutzpah underpinned by a comfort factor aimed squarely at the bulk of Oracle customers who are wedded to the company's database as the foundation upon which they operate. As one person said:
It's Oracle, what else do you expect?
Putting aside the irony of Ellison wielding the vendor lock in argument by parsing it to represent Oracle as presenting deployment choice, the real audience for Ellison's keynote are the bulk of customers who have yet to make a decision about which database workloads will move to a cloud based infrastructure. Layer on a healthy dose of how much Oracle is faster and cheaper than Amazon and you set up the stage for making an inevitable choice.
But then consider this exchange I had with communications chief Bob Evans about Oracle's R&D investments:
The subliminal message behind Ellison's 'speeds and feeds' at lower cost argument is clear: if Oracle customers want to move workloads to its infrastructure then it has the secure offering that first attracted the CIA to become an early Oracle customer back in the day and which can be offered in any flavor the customer wants.
In short, customers are assured of capability and support, the lynchpins of Oracle's business model. This is a wholly understandable defensive move for a company claiming 420,000 customers, something that the likes of Ben Thompson fails to grasp until the final gasp of an otherwise long winded and one sided discussion:
Oracle’s lock on its existing customers, including the vast majority of the largest companies and governments in the world, remains very strong. And to that end its strategy of basically replicating its on-premise business in the cloud (or even moving its cloud hardware on-premise) makes total sense; it’s the same sort of hybrid strategy that Microsoft is banking on. Give their similarly old-fashioned customers the benefit of reducing their capital expenditures (increasing their return on invested capital) and hopefully buy enough time to adapt to a new world where users actually matter and flexible and focused clouds are the best way to serve them.
In fairness to Thompson, he updates his analysis today in a private note by saying that Oracle appeals to the typical enterprise buyer.
Oracle's position is not without problem. I see four major issues.
Regardless of what Evans and others might say, the current offering is unquestionably a work in progress and while security is an obvious point to ram home for the enterprise buyer, Oracle has a lot of catching up to do to match Amazon's very flexible and growth based business model combined with Amazon's development spend rate which outstrips Oracle spend by an order of magnitude.
On the flip side, Amazon's broadscale offering works well for test and development environments but has yet to prove itself in at scale OLTP systems, a market which Oracle understands and serves extremely well. In short, the legacy buyer's needs are well met by Oracle but what of the more nimble start up that could be a future Oracle candidate? Another year or so will make that position clearer. But it is dependent upon matters it has yet to address.
My second concern with Oracle's infrastructure offering is around price. Colleague Vinnie Mirchandani likes that Oracle is attacking the IT operational cost base:
Those running PeopleSoft, JDE, EBS and countless other packages in the Oracle family use some form of internal data center or external hosting with IBM, HP and many others. That’s a captive market for Oracle IaaS to explore. Even more promising are SAP, Infor and countless other legacy customer bases who are today mostly using AWS or MS Azure in dev/test mode or as “burst” capacity.
I agree. But I can't see how Oracle makes money when it is pricing at 20% below the leader in a commodity market that has razor thin margins like the utility providers to which Ellison refers in his discussion.
Claimed gross margins of 61% sound good and well, but as we saw in the latest earnings release, sales and marketing cost are a significant component in a business model that needs both top and bottom line positives in order the keep the Wall Street masters onside. This is fundamentally different to Amazon that only has to show growth but which has the scale to demonstrate profitability.
My third concern relates to how the infrastructure message slots onto the overall application sales strategy.
Oracle's SaaS solutions have matured dramatically since 2015 and are far more attractive as suite offerings than they were this time last year. They are also attractive at the UX level, something I was not expecting. During the event, I met with customers who both 'get' the cloud story and are confident that Oracle's 'configure not customize' messaging provides a roadmap into which they can buy.
Based upon my quick fire Q&A with customers, I came away with the impression that apps customers are largely happy to compromise at the 80-85% functional level, provided they can slot in readily deliverable add-ins. That's a credible achievement from Oracle's perspective. However, it still lays open the question about how Oracle ties that completeness of functional story back to an infrastructure play that has yet to prove itself in the market.
My final concern relates to developers. Oracle's track record of supporting open source following its Sun and MySQL acquisitions has not been great. On Twitter, Matt Asay made the argument that:
@dahowlett Unfortunately he's making the argument to the CIOs who don't actually make the cloud decisions. Their developers do
— Matt Asay (@mjasay) September 20, 2016
This is true. When we are asked to support our stats, we point to the Line 1,2 people who subscribe to our content and almost never to CIOs because we know that the real technology decisions are taken at levels below the board. Increasingly, those same decisions are being taken at the operational department level where tech savvy leaders are deciding the future of application landscapes.
We can also point to the dedicated technical developer conferences that SAP, IBM and Microsoft put on as examples where the faithful are not only gathered but celebrated.
Oracle has not figured this out despite running Java One as an offsite event at the same time as OpenWorld. If it wants to win in infrastructure and platform then it needs to change the conversation by making developers first class event citizens.
While many colleagues concentrated their attention on the detail of Ellison's keynotes, I was more taken with the tone. His precise choice of words was brilliant, often turning past arguments from competitors upon themselves in a persuasive manner. Who can for example forget the words:
You can only have this if you are willing to pay less than Amazon
In that sense, Ellison weaves a tale of magic that is hard to dispel except by those who are accustomed to his approach. Challenges aside, I would be surprised if customers walked away anything other than comforted.