If every company has become a tech company, then The New York Times is a prime example. Although it launched as a print newspaper 171 years ago, in 1851, today that same company is very much a tech-driven media business.
While the transition has been ongoing to some extent for decades, the NYT brought in Jason Sobel in the summer of 2021 as chief technology officer to accelerate the transformation into a modern tech organization.
Sobel brings 15 years of pure engineering expertise to the job, including long stints at Airbnb and Facebook, where he helped to lead infrastructure. The Times brought in an engineer with this kind of expertise precisely because it needed someone to build the same kind of technology that was being built by the biggest tech companies.
In fact, Sobel said that he has found that the technical side of things isn’t all that different from his prior experience, except that everything he does is done in service of the editorial business.
“So it’s funny — it actually isn’t that different. I mean, the newsroom does make a difference in some ways. There’s obviously the strong editorial voice that’s always going to be important in the kinds of content we write and how we deliver it,” Sobel told TechCrunch.
But when it comes to building a technology organization, Sobel said what the paper of record is building actually feels quite familiar. “We have a cross-functional team of designers, engineers and product managers all working together to ship websites and apps and back-end tech. So I’ve been surprised at how similar it is,” he said.
Making all the pieces work
Sobel has to balance several different platforms, some of which are internally facing and some external. What’s more, he inherited some legacy technologies that go back decades, and he has to make all of this work together. That tech debt can be challenging when trying to modernize the tech stack.
“We have some systems that were initially built in the ’70s, for example, that have been updated and evolved and iterated. But Facebook, Airbnb didn’t exist [for decades] after that. So there’s [ … ] some more tech debt in certain areas than you might find [at newer companies],” he explained.
To combat that tech debt, he began to create a common set of services so teams could focus purely on building an application and not all of the adjacent pieces.
“One really big component of our strategy, and probably the biggest technology priority I’ve had, is around trying to centralize more of our core technology. I think the team had this idea before I joined but it’s probably one of the biggest problems facing us,” he said.
He attributes this issue to fragmentation as the teams moved from on-prem to the cloud. Lacking a centralized set of services, they were built from scratch multiple times for each new application. Sobel’s goal has been to reduce some of the redundancy by providing centralized services like authentication, messaging (push notifications and email) and payment services.
“So a lot of what our current strategy is, and that I’m personally really focused on, is trying to find a lot of those common patterns and figure out how to make them more shared across our team, which I think is a common narrative for growth-stage companies.”
Sobel said he saw a similar change at Airbnb, so moving his team to this approach felt familiar. “We actually went through the same thing that Airbnb, where we went from a monolith to a micro services approach, and we almost had the same problem. We had a proliferation of technologies and a proliferation of solutions, and then the CTO had the same idea of building a shared platform.”
Developers need common back-end tools, too
Sobel’s team, where the engineering practice is housed, has its own requirements for common services around containers, infrastructure and other nitty-gritty details that developers need to deal with, something Sobel wants to simplify as much as possible.
“Within our engineering team, for example, we built a container platform. And for people building a new service, we wanted them to use our container platform, rather than going and rolling their own thing. We found a lot of people who are excited about doing that this year. We’re trying to get to an early majority,” he said.
But it takes a level of change management to get people to buy into a new way of doing things and go with a centralized approach. “Some of the decisions we’re making in service of this strategy really fundamentally changes how people do their jobs. And so I think when I joined, the strategy was in place, but we needed to really land it and get started. And I think a lot of that involved change management.”
Sobel added, however, that it’s his job to get that buy-in and nudge more people toward using these centralized tools. “This platform is only valuable if we can get our our internal developers using it. And so we’re trying to push that next tranche of people onto it right now,” he said.
In terms of building internal systems, he takes advantage of open source when it makes sense but tends to favor a hybrid use case, moving between managed services and open source and adapting as needed.
“I think we’re in that sort of fairly common place now where our platform is a blend of managed services from cloud providers that we then build on top of and adapt for our use case, or we take an open source solution and do the same thing,” he said.
The team is still growing, and while he didn’t want to share exact numbers, he said his operation was second only to the newsroom. Because the NYT isn’t a pure tech company, he’s sometimes had to convince candidates for senior-level roles that working for a newspaper is as challenging an endeavor as any tech company — and actually rather similar to their previous jobs in terms of the work they’ll be doing.
“So that’s a big part of what I’m focused on, trying to get ahead of that question and how do we help people in a broader-base way understand that The Times does engineering much like any company folks would call ‘a tech company.’”
Ultimately, what draws people is the fact that his team is a huge contributor to the success of the entire company.
“We really believe it is probably one of the biggest differentiators for us as a business and as an organization. We think one of the reasons that business has done so well over the last few years is because we’re pretty far ahead of our peer news organizations in that way.”
Comment