What's the biggest blocker for legacy large companies' IT to switch to cloud technologies? originally appeared on Quora: the place to gain and share knowledge, empowering people to learn from others and better understand the world.
What's the biggest blocker for legacy large companies IT to switch to cloud technologies? It's pure and simple.
In tech companies, we recognize that there's what you'd call "cool tech" - all the heavy lifting machine learning and natural language processing and search and stuff that's not trivial to do and that really moves the world in various ways. And we recognize that there's what we call "undifferentiated IT services". You know, the stuff that is the same in company after company after company with only minor variations (user admins, file servers, email, EPR applications, etc.)
If you're a hot startup, the last thing you want to do is spend money on stuff that is not really moving the needle, but is sort of "ticket to play" stuff - you want to pull that burn to as close to zero as you can manage.
And, if you're a hot startup, you're trying to build a culture of "we're all in this together", and you want to value everyone equally. Except you've got this problem, because if you have in-house IT infrastructure, you now have all these people that you're not valuing the same, because... well, frankly, elitism. That PhD ML guy over in the corner? He doesn't want some network admin screwing around with his laptop to keep it conformant with IT best practices, he'll do that himself, if it gets done at all. And if you irritate him too much with that sort of thing, he'll move to the next hot startup,
So you quickly learn that you don't want any "undifferentiated IT services" inside your company, at all, if you can avoid it. It creates friction between those trying to serve the IT needs of the company as a whole, and those trying to move the needle. Bad juju, for sure.
About a decade ago, before AWS was a thing, there was a startup funded to build cloud infrastructure by a group of VC firms who were tired of paying for all this "undifferentiated services" nonsense in all of their portfolio companies. They figured that they might as well pay for it once, and then let their portfolio companies buy those services on a per-seat basis. That was visionary in 2005, revolutionary in 2007, and today? It's just how things are. If you've got any kind of tech startup, you actively plan on never owning infrastructure. It's a giant pain in the ass, and you can always buy it as a service from someone much cheaper than owning it yourself, and you can move orders of magnitude faster that way.
In non-tech companies, or tech companies where the "tech" is not computer/network/IT related, IT is handled differently. The company is completely dependent on IT, but that's not their core business. And the demands increase every year, and with that, the budget and relative power of the CIO and the IT team.
IT accreted over time, like a coral reef. There was a mainframe, then a cluster of mainframes, then PCs talking to mainframes, then PCs talking to servers talking to mainframes, then a move to Web 1.0, then Web 2.0, and that whole time, IT grew and grew, because there were only rarely wholesale moves to new tech, usually there were little initiatives that grew larger, while COBOL code from the '70s was still running somewhere deep down in there...
Add in modern cyberthreats, some malware, some ransomware, bring-your-own-device chaos, and these guys not only have outsized responsibility, but they also have outsized power.
So, there's empire building going on in IT-land. Few will admit to it, but it's a real issue. When it comes budget-time, you don't want to be the department that gets cuts, you want to be the department that grows. More budget, more and better tech, more staff.
It's sold on the basis of "controlling our destiny", meaning that my guys running my servers in my datacenter are somehow better than using public cloud infrastructure at 10% of the total cost to own, but it's empire building, pure and simple.
Sometimes it's driven by fear - fear that if you don't control everything, and something goes wrong, that you'll be fired for not having controlled everything (versus being able to say "that was Joe's job" and then firing Joe). Sometimes it's driven by will to power, the same thing you see in PTAs and neighborhood associations, the petty grab for control over the lives of others ("Heh, you may be the CEO, but you can't install anything on your laptop unless I say it's OK.") Sometimes it's a "career building" move - wanting to say "managed IT budget of $50M for global company" on your resume, and you're only at about $10M now...
But that's the number one reason that companies don't move to cloud.
Because moving to cloud disrupts that.
In one company that I re-structured, I turned a nearly $35M annual budget into a $7M annual budget, avoided a looming $50M tech refresh, shed over a hundred jobs and improved productivity and turn-around time on key informatics, by killing the in-house data centers and moving everything to the cloud. In most places, that would make you a hero; I knew going in that it would make me a pariah, at least in IT-land, but I didn't care, that was my job, and once it was done I could hire a CIO that understood the new paradigm and leave him to manage the process of re-aligning the staff with reality - or replacing them if they didn't get on board.
That opportunity exists in nearly every non-tech-centric company in the world.
The problem is, you're fighting entrenched hegemony - the CEO is afraid that if the CIO screws up, they're both out of a job. Taking risks is part of the fabric of existence in startups; in running enterprises, that's a much less desirable thing, and is seldom done, unless something forces it.
This question originally appeared on Quora - the place to gain and share knowledge, empowering people to learn from others and better understand the world. You can follow Quora on Twitter, Facebook, and Google+. More questions: