In a startup there is no more important task than choosing what to do. Even a minor improvement in focus can yield huge results. With better decisions you can get away with a smaller team, less overhead, and less outside funding.
By most measures 2013 was a huge success for our startup. All our core metrics improved substantially, but our product didn’t change much. We were getting bogged down with small features, maintenance and bug fixes but failed to deliver big improvements to the user experience.
We keep track of our work using two lists: One for big ideas that we want to build one day and a second “to-do list” for items that are scheduled to be worked on. I would prioritize the to-do list each week, trying to insert big ideas into the workflow. Despite my efforts, everyday a new bug or small feature would suddenly become a priority. Those little things consumed our development time and the big ideas rarely made it in.
To cut through the noise we tried scheduling a one week sprint on a big idea. We picked a project on the Monday and delivered the feature by the Friday afternoon. For the first time in a while, we managed to get one of our big ideas out the door.
Feeling energized by that experience, we took on a more ambitious project for our second sprint. By the Friday small stuff had crept back onto our to-do list and we were only half finished our big idea. Argh!
It seemed like sprints had potential, but there was something missing in the way we chose our projects and defined their scope. I wanted a new process that would do three things:
• Ensure we consistently worked on big ideas with big impact
• Limit scope creep and keep the product simple
• Provide time to cover inevitable bug fixes, maintenance & small features
We started off the year by deleting the to-do list completely. We kept our list of big ideas, but all the little fixes & features were gone. We were apprehensive about loosing all those notes, but I was confident that this purge was necessary to get us on a better path. Once it was done it was incredibly liberating. Without the weight of the to-do list we could do anything. The whole year opened up in front of us.
We started by creating a loose roadmap for the year. Inspired by Ian McAllister’s post on roadmapping with themes, rather than projects, I took our goals for the year and assigned a corresponding theme to each. I then lined up the themes across each month on the calendar. A month is enough time to get deep into a subject and really make meaningful change. But it’s not so long that we would have to worry about ignoring other parts of the business. The roadmap also allowed us to think though how different themes would fit together over the course of the year.
Now every Monday morning the team rolls in with no pre determined list. Instead, everyone has to pitch their idea for what we should work on that week. Ideas have to be on theme, but there is no other restriction on what qualifies. We each score ideas by two categories: simplicity & impact. The idea with the most points is what gets worked on. Full democracy.
Scoring on impact encourages team members to think about what will really make a difference to the business. Scoring on simplicity ensures that we take on projects that have minimal ambiguity, are achievable within the week, and minimize the complexity of the application.
Each Monday we stop work on the prior week’s project. If there are features that we didn’t get to, they are either deleted or put in the icebox for the end of the month. Having that discipline to stop work on Monday and question the impact of what we’re working on ensures that we aren’t held back by sunk costs.
Over the course of the month we keep track of all the small stuff in a separate list, the icebox. By the fourth week it will have filled up with all sorts of minor fixes that now need to be addressed. So we take the last week to catch up on small stuff. Some teams set aside one day a week for bug fixes, but we’ve found that having a full week for bugs and small stuff works well because you can take on mid sized projects that are annoying for users but not huge priorities.
On the last Friday of the month the to-do list gets purged. We look back at the results of our projects to see if they were a pass or fail. We also discuss our process and how it might be improved. Then we delete everything in the icebox and start completely fresh on a new theme the following month.
With this new process in place we work on big ideas every week, and it has made a huge difference. Cutting off our projects at the end of each week and purging the ice-box at the end of the month has forced us to consider what is important now and prevented us from doing things that might have been important six months ago, but are no longer relevant.
Pitching to the team has forced us to refine our ideas and discourages people from spitballing without a supporting argument. The two dimensions, simplicity & impact not only encourage people to select ideas along those dimensions, but also to optimize for them. How could an idea be more focused, more impactful or more achievable? That’s been an awesome and unexpected result. Not only is our selection better, but so are the ideas themselves.
With this new process in place SpaceList is getting noticeably better each week. We are no longer behind the curve, we’re ahead of it. More that that, we get into the office on Monday morning knowing that we are going to work on something that matters.