Don't Build It Unless You Have To

My new motto for all new ventures is: If I don't have to build it, I won't. This is a marked change both in my life path and the advice I've given others. In the past, I felt that without building most elements of a digital project or a workflow from scratch, you couldn't reach something close enough to your aims to achieve those goals. Custom work, typically involving a lot of coding, was the only way.

I've gotten past that. The Magazine was a grand experiment in building it from scratch, and I credit Marco Arment tremendously for putting in the time and effort to make it happen. It was the only reasonable approach in 2012 to produce a born-digital and digital-only publication distributed to mobile devices. As I wrote at Six Colors, app ecosystems used to promise and deliver almost everything you needed for a publication; now, they promise more than they deliver, though there are still advantages.

The Magazine had these built-from-scratch properties:

  • A custom app, which was perceived (incorrectly) as iPad only
  • A built-from-scratch web site (turned into templates)
  • A custom back-end for both iOS and web app interaction
  • An in-house account management system to integrate iOS subscribers (by receipt) and Web access, as well as allow Web subscribers to use the iOS app
  • A custom Apple Push Notification (APN) system
  • Initially accepted payment solely via iTunes; then added custom ecommerce handling
  • Did not offer integration in the app to allow subscribers to add themselves to a mailing list

This all made sense in 2012 because:

  • There were no mature periodical platforms. Now there are several. (I picked TypeEngine for our 2.0 update this last summer.)
  • WordPress was the only reasonable and mature offering for hosting a web site with the complexity of what we needed in 2012, but it still wouldn't have been the right choice then. Now there's Squarespace, among others, which are much more sophisticated, even without being able to run custom PHP or the like.
  • Marco is a PHP and iOS programmer: he didn't have to hire in any expertise except in the user-interface design.

The trouble with custom everything, even if you are the person writing the code and are proficient (I can program perl and PHP but not Objective C), is that every single change you make or feature improvement you need is a slog. You're the only one who can do it. If you job out some parts of what you do for custom work, you have to manage those projects and get other people to conform to your needs, while they have other priorities they're juggling.

I have an increasingly well-formed idea for a new publication that I may launch in late winter. For this project, I swear to the heavens above, I'm going to stitch together everything I need from existing components, and only write the glue to bind them. Squarespace offers a lot of glue in its setup: linking in Stripe, Mailchimp, Disqus, and many other services by just popping in those other servcies' API keys.

  • If there is an app component, it will be a publishing platform that I license or to which I subscribe, not an app I commission.
  • Any content available through a platform will also be available in ebook format.
  • Web hosting will be on Squarespace.
  • I will not write a line of ecommerce code, but design the project around the capabilities of existing integration in Squarespace or another system that I can link to the web site.
  • I will not build an account-management system.
  • An email list (using Mailchimp) will be a fundamental part of communication.
  • It will not be beholden to Apple or any monolithic company for funding, ongoing subscription revenue, or feature approval.

I will focus all of my efforts on editorial, marketing, and design. Am I being naive to think that what I need is available? Not really. I've tested every one of the elements I mention in the last bullet list in isolation. The trick is making sure I can create an integrated whole. I believe it's possible.