Oracle has a Mark Zuckerberg problem


Mark Zuckerberg

Screenshot via MSQRD

Mark Zuckerberg trying out MSQRD

In 2013's "Iron Man 3," Tony Stark's digital butler sidekick JARVIS uses the Oracle Cloud to reconstruct a crime scene and track the villainous Mandarin. It was a weird bit of product placement.


In 2016, though, when Facebook CEO Mark Zuckerberg announced he was building his own version of JARVIS to power his smart home, he decided to start from scratch and write it himself.

This, in a nutshell, is the problem with Larry Ellison's bold claim that the newly upgraded Oracle Cloud will end Amazon's runaway dominance of the cloud computing market with an alternative that's both faster and cheaper than the $10 billion Amazon Web Services juggernaut.

It's likely that the so-called "next generation" Oracle Cloud will capture a lot of existing Oracle customers as they move to the cloud computing model, where you can rent functionally unlimited supercomputing power for pennies per hour.

But the next generation of young, talented programmers - the Mark Zuckerbergs of the world - have very little incentive to jump on board the Oracle bandwagon, and every reason to keep on with Amazon Web Services, where they can do things their own way.


Power isn't everything

In a new edition of his Stratechery newsletter, analyst Ben Thompson lays it out plainly: The Oracle Cloud's major advantage, as Ellison himself says, is at the high end of the power spectrum. If you're running super-intensive computing applications, Oracle Cloud does indeed seem to be faster and cheaper.

That's not the market Amazon is going for, though. The thing that's set Amazon Web Services apart isn't its performance or its price; it's the fact that it's a platform on which to grow. You can start small, with one or two servers, add capacity as you grow, and then instantly tag in new services like databases or video hosting or even artificial intelligence, you can just tag it in as you go.

Writes Thompson: "Sure, it can get expensive - a common myth is that AWS is winning on price, but actually Amazon is among the more expensive options - but how much is it worth to have exactly what you need when you need it?"

In other words, Oracle Cloud is focused on power for the high end, but has less to offer around the low end. Amazon's strength is owning that low-end, then upselling more services a la carte as companies grow, change, or move more workloads to the cloud.

Have it your way

And speaking of a la carte, that pricing model has long been the key to Amazon's success.


While Amazon's biggest customers will pay a contracted rate, the vast majority of Amazon Web Services customers enjoy a pay-as-you-go model that reduces upfront investment and gives room to experiment and grow. It makes it easier to sell products straight to the software developer, bypassing the traditional IT department procurement process. Under this model, developers can just go online and get what they need, as they need it.

Meanwhile, Oracle takes a very old-school approach to selling big chunks of products all at once through the CIO's office. It's not the way that developers enjoy doing business, and thanks to Amazon and the business models it's inspired, it's not something they have to put up with.

"Developers today have choices, and want to work with companies that are innovative, transparent, and friendly, and with platforms that are easy-to-use. For legacy vendors built on lock-in, closed systems, complex setups requiring long contracts and professional services - that would require a complete DNA swap," says Gleb Budman, CEO of Backblaze, which competes with Amazon in providing cloud storage.

Not doomed

Oracle will do just fine. For its biggest customers, the Oracle Cloud is a no brainer. Specific applications for stuff like sales and marketing can move up to the Oracle Cloud and save them some money, while the higher-end cloud infrastructure capacity in question is perfect for hosting large databases.

Oracles moves will also put pressure on its long-time rivals at Microsoft, who often compete for the same large enterprise companies. If Oracle's making it easier for its existing customers to move up to the Oracle Cloud, it's that much harder for Microsoft to poach them to its own Microsoft Azure.


And with the sheer amount of well-paid cloud talent at work on the Oracle Cloud, there's little doubt that it works as advertised. The technology, as presented, is impressive.

But almost any company starting today - the next Facebook or Dropbox (which started on Amazon and took years to move to its own data centers) or Snapchat (which runs on Google's cloud) or whatever - has few reasons to pick Oracle. Larry Ellison is correct that in many ways, Oracle's technology is way ahead of Amazon's, but that gap is narrowing faster than he thinks. And the company just isn't well-suited to competing on Amazon's turf.

There's always a chance. Oracle has lots of cash in the bank, they're investing heavily in research and development, and the money that comes in from older customers upgrading is going to be a major windfall. If Oracle can successfully invest in a true reinvention of itself, maybe it can hang. For now, it's going to be slow going.

Disclosure: Jeff Bezos is an investor in Business Insider through hispersonal investment company Bezos Expeditions.

NOW WATCH: I tried sending something to an Amazon Locker and it was a terrible experience