"Running a data center is a lot like running a business," he adds. "Even having an in-house data center didn't provide us with the agility and flexibility that we needed as we went down this digital path. So we decided the best option for us, really, was moving to a cloud solution."
Stopping Costs of Failing Projects Was Big Selling Point
One of the big selling points at Condé Nast was the capability to shut down a runaway project and remove the cost immediately.
"Let's say you're on a runaway project; there is so much inertia not to stop it," Simon says. "We've had a few of those. When I first came on, I shut down five or six projects that had been going on for five to seven years with no end in sight. If something doesn't work, you shut it down and the cost goes away."
"It took time to understand the economics," Simon adds. "But it's not a one-way street. You can always go back if you need to. But in the end, we are not in the business of maintenance. We're in the business of rapid change."
Once he got the green light, Simon and his team hit the ground running. They reviewed a slew of service providers for capability, functionality, flexibility and cost. On every front, Simon says, Amazon Web Services topped the competition.
It also didn't hurt that Amazon has long been Condé Nast's ally in the digital trenches — Simon notes that Amazon was an important partner when Condé Nast fought Apple to keep control of its customer information for subscriptions via the Apple App Store. Control of that sort of customer data is one of the advantages that oldline print publishers can bring to bear as they move into digital and begin mapping that data to web and smartphone behavior.
Short Deadline for Migration Ensures Success
Simon set his team a three-month deadline for the migration to AWS.
"Once we chose AWS, the whole planning and execution took three months," Simon says. "We didn't have too much time to sit and argue over the house. We were going to get this done in the most cleanline fashion we could."
Simon set such a strict deadline because he felt it was the best way to ensure success.
"Everything I've done with a tight timeframe has worked," he says. "Everything I've tried to do with a 'loosey-goosey timeframe has always failed. People will say it's the prudent thing to hedge your bets, but if that's how you're going to do it you might as well go home. I already had a backup if it failed; it's called my data center."
Sign up for Computerworld eNewsletters.