NewsTilt

» see newstilt.com as it lived, on the wayback machine

~ the eulogy ~

NewsTilt was alive for ten weeks in 2010 and I am not going to pretend I noticed it at the time. Almost nobody did, which was the problem it died of. I came to it the way most people who know the name came to it, backwards, through the document its founder wrote afterwards.

Paul Biggar and Nathan Chong were computer scientists who started the company in October 2009 and took it through Y Combinator. The idea was aimed at a real grievance. Independent journalists were producing good work and had no way to build an audience that belonged to them rather than to whatever publication had bought the piece. NewsTilt would vet the writers and host their work under their own names, taking twenty percent of whatever came in.

It launched on 13 April 2010 and immediately did not work. Traffic never arrived. The commenting system required a Facebook login, which in 2010 was read by a chunk of the audience as a privacy demand rather than a convenience. The journalists who had signed on watched their numbers and understood fairly quickly that the audience they had been promised was not going to show up.

Biggar defended his doctoral thesis and got married in the weeks right after launch, which is not a schedule that leaves room for rescuing a failing company. He and Chong stopped being able to talk to each other productively. On 26 June, ten weeks in, they shut it down.

~ the dossier ~

~ what they got wrong ~

The diagnosis in Biggar's own account is that they built for a profession neither of them had worked in. The specific version of that mistake is precise enough to be useful: they did not understand that journalists want editors. Both founders came from software, where the writer of the thing is also the person who ships it, and they built a platform on that assumption. Writers who had spent careers with someone reading their copy before it went out found a tool that handed them a publish button and called it freedom.

Chong's farewell email to the journalists said it more bluntly than most shutdown notices manage. They should never have made promises about building anyone's online brand or delivering large amounts of traffic. That is the sentence, and it applies to a very large share of the products that have ever been pitched to writers.

~ the funny part ~

Biggar wrote the postmortem in September 2010 and it went further than the product had ever gone. Gigaom covered it. So did Bloomberg. A two-month failure with no users became one of the more widely circulated startup autopsies of that year, because it named the founder relationship as a cause instead of hiding behind market conditions, and because it was specific about which promises had been dishonest rather than merely optimistic.

As an engineer I find the shape of it entirely legible. Two technical people look at an industry from outside and see an obviously broken workflow. They build the fix without ever asking the people inside why the workflow is the way it is. I have watched that happen in domains I do know. The difference is that Biggar wrote it down while it was still embarrassing.

~ the inheritance ~

Biggar went to Mozilla and then founded CircleCI, which became a large and successful company. Darklang came after that. Failing this publicly does not appear to have cost him anything.

The problem NewsTilt was aimed at never got solved, it just got renamed. Substack and the newsletter economy behind it are selling the same thing fifteen years on: your audience and your name on it, with a platform taking a percentage. What they had that NewsTilt did not is a distribution channel that already worked, which turns out to have been the entire product all along.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »