I love using pencils. There’s something about the feel of the graphite on a blank page. The pencil is a constantly changing writing device—both the width of the graphite path that follows, and the length of the pencil as it’s used. Most of the time it isn’t because of the ability to erase and change what was already laid out, but that’s a pretty significant feature too.

The past few weeks, we’ve had to use the eraser a bit. We were full steam ahead, ready to roll out Cardsmith to the world. It was beautifully usable, details had been nailed, and seemingly every ‘user path’ tested. We started signing up paying users—founding members—and getting them into the product. And then we started seeing some issues. For months we had been using Cardsmith internally and had a great group of Alpha Testers. We had experienced zero data loss and were using it daily for every aspect of our business and personal lives. We had big boards, small boards, tall boards, wide boards. We had boards with images and boards with text. We had boards with every color of cards. What we never had were boards that lost any data. Until we started rolling out.

It felt like we hit a brick wall. We tried some initial patches, but discovered that the issue was deeper than we could easily solve. So we used our eraser a bit. We backed up a few steps and made some changes. While we wanted to have thousands of happy users by now, we realized that our long-term success was important enough to not mess this up. Shortcuts aren’t our style and while it was a bit disheartening initially, we’re excited about the possibilities that are ahead as we’ve re-written some of the details of our plan. It’s been more than a few weeks that we’ve been saying “we’re still a few weeks out from launch,” but we are. Maybe three, maybe four or five.

We’re getting there, following the undulating graphite path as our pencil shortens and progress gets made toward an even better, more solid product.