Better Software - January 2009 - (Page 103) The Last Word Are Your Pants on Fire, or Do You Suffer from Split Focus? by Johanna Rothman Imagine you’re working on a project. You’re dealing with risks and making technical decisions—pretty much humming along. The project isn’t easy, but you’re making progress. One day, you arrive at work and your boss says, “Stop working on that project. Work on this one, instead.” You do. A week later, the same thing happens. Welcome to the Pants on Fire schedule game. Pants on Fire is bad enough, but some managers don’t even let you work on just one project at a time. I had a boss who said, “I’d like you to spend 50 percent of your time on project A, 35 percent of your time on project B, and 15 percent of your time on project C.” You and I both know there’s no way you’re making progress on projects B or C, but they do successfully split your focus off project A. The multitasking you’re doing is the Split Focus schedule game. Split Focus and Pants on Fire have similar causes: Your managers can’t or won’t decide which project is most important right now. Making those decisions is tough, and some managers don’t know how to decide. People who have been working in the software industry for more than three weeks have probably encountered these two schedule games. Both of these games cause context-switching for the project team members—including the project manager—and will slow your projects, sometimes to a crawl. What should you do if you’re a project manager or technical lead and you realize management is causing most of the problems on your project? Prayer might be helpful but is not enough. Here are some actions you can take. doing in a project portfolio. such a difficult time selecting Making those Once you have a portfolio, the most important project you can analyze it to see and keeping people on it is which projects are at which that they don’t know when decisions is priority. you’ll be done with your To build a portfolio, project. With timeboxed iterations, you can show them tough, and some manag- first collect the list of all the work for all the people who how far along you are at the are being shuffled from end of the iteration, and you ers don’t know one project to another. have a better chance of preRemember, it’s the list of dicting when you’ll be done. project work, ongoing Say your organization how to decide. work, periodic work, ad needs to release something in six months, but you won’t be done hoc work, and management work. Organize all that work in a grid, for eighteen months. You might be able either to release something from with months across the top and the list this project in six months or to put this of projects down the side. Use the estiproject on hold until you’ve completed mated start and end dates for each piece of work, so it’s easy to see when you another project in six months or fewer. think the work will start and end. Once you’ve got the list, try to rank Know What “Done” Means Make sure you’re doing the minimum the projects. That means assigning a required. Sometimes your managers will unique number—1, 2, 3, 4, and so use Pants on Fire to move people off on—to each project. The highest-ranked one project that they think is complete projects are the most valuable to the enough. If you haven’t defined release organization. Discussing a ranked list criteria, make sure everyone understands with your manager might help you have a more productive conversation than if what “done” means for this system. you just discuss a list of projects. Estimate with Ranges Start with an Agile Approach If you’re not already working in short (one- to two-week), timeboxed iterations, implementing by feature and integrating as you go, start that now. One of the reasons your managers might have Sometimes managers invoke Pants on Fire or Split Focus because the team missed its estimate for project completion. Instead of only providing one date for project completion, try providing a range or a confidence level with different dates. Say to your manager, “I have a 70 percent confidence we can meet June 30, and a greater than 95 percent confidence we can meet September 15,” and you’ll have a different conversation about when to stop this project and start another one. Qualitatively Evaluate Each Project If you don’t know the value of each piece of work, use these qualitative questions to evaluate them: • How does this project fit in with all the others? • What is the strategic reason for this project? • Is there a tactical gain from completing this project? • To make this project successful, are we ready to adequately fund it and staff it? • Do we know what success looks like for this project? Sometimes a qualitative evaluation isn’t enough, so you might need to use numbers to evaluate the projects. BETTER SOFTWARE Build a Project Portfolio If you’ve already taken these steps and your management still wants to have your project team multitask or switch projects frequently, then it’s time to show management what everyone is www.StickyMinds.com JANUARY/FEBRUARY 2009 103 http://www.StickyMinds.com
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.