Paul Ford is the co-founder of Postlight, a digital product studio based in NYC. Before starting his own company, he covered tech for many well-known publications like The New Yorker, Bloomberg Businessweek, Slate, and more. Ford is also a programmer who created SavePublishing and AnxietyBox. Based on his passion for coding, Ford wrote a massive 38,000 piece explaining what programming is for non-programmers, called “What is Code?”
We chatted with Ford about smarter content management systems, his projects with Vice and Bloomberg, and the future of media.
— Please tell us about your most recent project, Bloomberg Lens.
First, let me take a step back and share one of the big projects that came out of Postlight Labs: Mercury. It’s a Chrome plugin that parses web pages and makes them more easily readable. It’s similar to an earlier product that my business partner Rich Ziade created, called Readability. Underneath Mercury, there’s a platform that can do a lot of things: it has APIs, you can use it to convert content to be AMP-ready for better mobile reading. As we’ve been working more and more on Mercury, we’ve been thinking of what else we can do with it.
One of these things is Bloomberg Lens. It’s a little plugin that gives you information from Bloomberg about stock symbols and people in news articles you read across the web. The goal is to be non-invasive, just to let people know that there’s more information they can access.
Bloomberg Lens reveals data and news related to the companies and people you’re reading about.
We all knew Bloomberg had unbelievable data, but it’s been locked away in their terminal for years, and we paid $20,000 a year to use one of these terminals. But the Bloomberg APIs are freely available—they are giving away tons of data that anybody can access. So there’s all this intelligence that’s locked away on pages that could be extracted and used to help people find new information and engage more. Over time, we are hoping to get all this data and make it available to people who are creating new content. I’d love to see this stuff start becoming available at the CMS level and making every webpage really, really smart. Then, when you sit down to write an article, instead of having to pull up a bunch of web windows and cutting and pasting information, you could have information easily accessible to the writer or creator.
This project reminds me of tools of the early web. I grew up on the web that was very much about media and publishing. Over the last 10-15 years focus has shifted—now that the web has ways to deploy and ship applications, even content-rich sites like Facebook and Twitter aren’t really media sites, but are apps that happen to have an enormous amount of media flowing through them. The culture of the web has gotten to the point where people are very focused on creating these apps.
But the interesting thing about media is that—as opposed to an app that you’re lucky if a thousand people download—if you happen to have a hit article tens of thousand or hundreds of thousand or sometimes millions people can easily read it, even if nobody has heard of publication before. So what we are trying to do with Mercury is attempt to bring some of the intelligence of apps back to the media experience, connecting pages to other APIs and services, thinking about ways that paragraphs and sentences can be enhanced. We’re back where we started. There were sidebars all over the web in 90s and early 2000s, there’s some irony there—maybe we’re just old-fashioned, trying to make it the way it was.
— I’m really curious what have you learned during this year and a half working on Postlight. Which projects did you enjoy the most, which lessons have you learned?
First, I thought starting a company like this meant that I could say “let’s get something done” and it would get done. And that’s not how human beings operate at all. You have to point them in the right direction, and I think that has been the big lesson.
I also used to be a bad-mouthed, opinionated person in the media and then suddenly I found myself in this capitalist enterprise where I have to keep my mouth shut.
There have been a lot of things I wish I’d have known. Every day is a new and sometimes painful surprise, not necessarily because of other people, but because of my assumptions about what would it take to run a business. And it’s very hard to get a single webpage up in the world, so that part has been a lot to learn. The web is complicated, it’s highly collaborative, and it’s challenging to operate when there is such a big mess of platforms and JavaScript frameworks and approaches.
Ultimately, what I’ve really learned is that I’m glad I’ve chosen my partners well. Having someone who has deep experience—well beyond mine—in running an agency and selling services meant that I didn’t have to invent a lot of things.
It also turned out that the things that you worry about before you make stuff like this—I don’t have the right clothes, I’m too fat, or too old—all these things that you’re completely sure will keep you from making a deal are completely irrelevant. If people think you can get the work done for them, they don’t care about anything else.
— Which of your projects did you enjoy the most?
One of the projects that really stands out was for the People Magazine division of Time Inc. They wanted to build a new property focused on following social media stars called Instant.me. Neil Renicker, a very talented designer and engineer, ran that. He worked with our design team, but he added and implemented a lot, even building a custom CMS. Ultimately they started a new business from it. I’m very proud whenever we can start something new, not just fix something. To me if that’s one true satisfaction of doing this professionally instead of being a writer—you can make an opportunity for a creative person to do something.
We also worked on two projects for Vice. First, a large part of Postlight has been working closely with their product leadership to make a new CMS that powers most of their brands globally. I’ve worked on consolidating many media companies’ content management systems and they tend to go awry because they take so long. So all of the different media verticals and brands go, “Ok, screw it, I need to do it on WordPress.” But at Vice we actually pulled it off.
Second, while that was being built, Vice had a new news program with HBO called Vice News Tonight. They needed to create a mobile-first website for it that is hosted on WordPress. All the headlines had to be animated, and all sorts of complicated interactions needed to happen. So we hacked hell out of WordPress, just made it do crazy things to make it more dynamic. It was a big install, and we got it up quickly.
Being able to do multiple projects like this inside the same giant media organization makes me feel that we are little more grown up.
— When you should build your own CMS and when you should use WordPress templates? How do you choose when you build something new?
I think there is no truly wrong answer. It comes down to what the culture of the team is, and by culture I don’t mean just the work culture, but how you do ad sales, where you are globally, where you are going to be in 5 years, etc. Vice is a great example, because it’s an organization that wants a certain level of control over the experience. They are global, and different publications have different looks. They needed to create a space where they could easily expand and where they had real control over hosting in order to scale the organization.
When you choose WordPress as a platform, you are saving yourselves a year of development time and you are gaining an enormous amount of knowledge for free. And you get the tools—all the plugins and everything—to make it fast and powerful. You’re not gonna have to make a lot of choices, and you can move forward quickly. So if your goal is to put a standard page on the internet with standard advertising, you should just take what’s there and build on top of it.
There’s certain scale where off-the-shelf answers don’t really work anymore. You can think about it in terms of efficiencies. If you are 10-person team, that a 5% improvement in efficiency from a custom CMS does not mean much. You need some real boost in order to justify doing something big with software. But if you’re multi-thousand person organization, a 5% improvement in efficiency is huge and it’s absolutely worth your time to put people to work making a better solution. Overall, if the CMS is not the most important tool at your company, and you are relatively of small size, you should probably choose WordPress. But if there are genuine efficiencies to be found if you change the way you work and you want more control, it might make more sense to build or license something.
If people think you can get the work done for them, they don’t care about anything else
— Why did you decide use Medium for the Track Changes platform?
We reached a point where all of the podcasts and newsletter staff came together in a moment of panic because we were just not getting enough money to get going. So we just started recording and I started sending newsletters.
I’ve been an adviser at Medium for years, and I also have a large following on Medium, so I decided use it. Longer term it’s tricky, because a lot of the SEO goes back towards Medium. As a web company, I’m very conscious of the fact that it’s there rather than on our own site. But Medium really got us in front of people in a way that other platforms couldn’t.
Before Postlight, for pieces I wrote to feel like a success I needed to have hundreds of thousands of people reading them. With Track Changes, I realized that if I reach a couple thousand of people and one of them sends me an email saying they’d like to work with us, it’s incredibly successful. In the podcasts we talk about really technical, abstract things and they might not be that exciting to everybody, but for the 100 people who get it it’s extremely valuable.
— So what do you think about the recent news about Medium switching business models?
I can’t say much about it as an advisor, and I wasn’t involved in a decision. I’m gonna hold out and see what happens. I feel it’s good for someone to say out loud that the advertising model is not great, that it is failing, complicated, and insane. The publisher can say they are helping the marketer find the right audience, but you can’t really guarantee that they’ll deliver that audience. But then marketers kind of don’t have a choice if they want to boost their signal, because for whatever reason people will only read so much about your new shoe or your new prescription, it has to be aligned to it in other media experience that they want to participate in. On the other hand, the poor publisher gets basically no money from the marketer.
So given we are in this environment I think it makes sense to start thinking really hard about the unthinkable which is that we may not have a future with advertising, with media. We probably do, it’s always gonna be part of it because advertisers don’t have distribution. But in terms of blowing up marketing and blowing up advertising model I’m glad to see anything new.
— Tech companies are trying to create their own media experience—Facebook’s Instant Articles, for example—so brands don’t need a CMS. What do you think about that?
I think it’s kind of a mess. Media drives tech crazy, because it does not make sense as a business and yet it’s incredibly powerful. It’s all very arbitrary and old-world and irrational. It’s kind of a bizarre match up because The New York Times is valued at around $1-2 billion, and Facebook is valued at like 100 times that, but it’s not 100 times more powerful. When you put these two worlds together, it just does not make sense.
At the same time, Netflix is fascinating. They became a major network, they hired their own people, taking them from HBO, ESPN, and all over the place. They got the platform part right, making it so you can only watch Netflix video on their platform. Just thinking from an infrastructural level, it’s very much what Google does, but then they make first-class content that people spent millions of hours consuming, and there are no ads. So what would scare the hell out of me is Facebook buying Netflix. That’s the apocalypse. Now Facebook has—not just an ability to show you your friend’s cat—but the ability to hold your attention for hours through TV series. And right now it can’t do that.
For the first 20 years of my career people have been holding onto the idea that it will all somehow revert to normal. Facebook will disappear, people will subscribe to magazines again, and they will only subscribe to 4, and once a month they will go to a coffee shop and read those magazines and talk about it. I think that fantasy is dead. The new generation came along, and the new generation likes their phones. They like to talk, and they are also exhausted by social media. I have young kids, and they are less interested in my phone than I am. They just see it as part of ambient environment.
— You build AnxietyBox, a tool that sends you emails about your anxieties. Does it somehow reflect your thoughts on the future?
AnxietyBox just sent very mean emails all the time. In America, and especially in the tech industry, you can get very stuck on the idea of “oh my god, I need to make a billion dollars and anything less than that won’t be successful.” But that success has nothing to do with happiness. Making the minibot that made fun of me was good and kept my head straight.
The project ended up on “This American Life,” so I set up a Google Form and asked people to list their anxieties with the idea that if I ever launched it I would use that to start the app. Now I have 8,000 people in that form. And they listed all their greatest fears. Everyone is exactly the same: they are terrified of dying alone, that people will figure out what a fraud they are, that nobody loves them, etc. I love this aspect of this project—it removes human drama and you don’t get to feel very special anymore.
— What are you planning to do in this difficult media environment?
The thing that I see myself doing in the future—which actually ties back to the idea of Postlight—is having a platform that enhances pages. On my off hours I continue to work on experimental content management systems because I can’t leave it alone. I believe that the internet is really good when you make things that help people make new things. Most of what I see enhancing CMS’ now are things that will drive traffic, making things that are more meaningful to Google or to other search engines. I’d love to see some of that intelligence focused on how we can make writers and editors more powerful. How can we give them authority and not just show them statistics and analytics, but help make it a better article, better video, or better podcast?
I’m convinced that given the entirety of the internet and all the resources we have that we could do at least a little bit more. It wouldn’t have a direct impact in the same way—you’d have to figure out a way to measure it in order to convince people to do it. Certainly nobody’s gonna run up give me a lot of money to pursue it, but that’s ok—it can be a side project while we keep the company going.
