A Shift To Web-First Development

As most of you know I’ve been “mobile first” my entire career. We developed our first apps for Newton and PalmPilot in 1997, although Apple cancelled the Newton before we shipped an app for it. We then developed for Windows Mobile and Windows, although our Windows version was designed for the pen-based tablets, not the desktop OS. This was followed by BlackBerry, iOS, and Android.

When we started developing Equals we started it as an iOS application. Multiple rounds of prototype development were done on iOS, where the web served primarily as connecting tissue, syncing templates across devices and giving our customers a read-only view of their notes so they could be followed.

I was always uneasy with this approach. There are significant problems starting a service on mobile devices. Here’s a few of them:

  1. Gives us massive scale immediately
  2. Requires a high level of polish
  3. Long release cycles
  4. Connects us only indirectly to the customer
  5. Offers only vanity metrics

When starting a new service, all of these are required in reverse. We can’t handle massive scale because we don’t know how or where to scale yet. Plus, we need to control who uses the app. We want a rough product to launch that can be revised quickly and easily. We need tons of customer touch points so we know whether we are on the right path. We need in-depth data and knowledge to refine our metrics, not vanity metrics such as downloads that tell us next to nothing.

Last year I fully realized the folly of my ways and we started making the shift to a “web first” approach. We had minimal skills here though. As of last summer I had never written more than three or four lines of JavaScript. I knew what responsive design was but had never thought about how to implement it. We did (and do) know Rails, CSS and HTML basics, so we weren’t completely starting from scratch, but all the same I knew it would be a difficult but necessary transition.

I lined up contract gigs that taught us HTML5 and got to use responsive design for the first time. We refined our skills for most of the year, iterating over development projects and working on Equals between the contract gaps. We have the skills we need now and are full-time on Equals. We will have a web version that runs on various desktop and mobile browsers available before April is out.

We still have plans for mobile-native versions of course, but this gives us the flexibility to build at our own pace, learning a ton, refining the features, and then make the best possible mobile apps we can when ready.