Menu ▼

Planning and deconstructing a project

Find a set of steps to get to the desired outcome. If a step is not clear and actionable with clear inputs and outputs, break it down further.

There are three types of steps:

  1. You do the work without oversight.
  2. You do the work, but someone else has some information or knowledge you need, or they need to verify your work.
  3. Someone else needs to do the work.

The first type should be the easiest to estimate in terms of time and complexity, and since everything is in your control, you can do it anytime. The last type is the hardest because you depend entirely on the motivation, skill, and schedule of others.

Find the limiting step. The limiting step is one that takes the longest and one you can’t break down into smaller steps, so your project will take at least that long if you’re able to do everything in parallel.

When ordering the steps, start with the limiting step as soon as possible, schedule the work others have to do early, and be most flexible with the work you know you can perform fast. It won’t always be easy to plan this way because many steps are sequential and dependent, but it’s the ideal you should strive for if you want to make the project as short as possible.

It’s prudent to monitor the progress of your project. However, every monitoring system or process has a cost. There aren’t any hard rules when and how much monitoring to build in—it’s a judgment call on how much to invest and slow down a project to catch mistakes early, which could be, if not caught, catastrophic later on.

Previous blog post:
Reading on the web became hell

Stay up to date:
Email · RSS feed · LinkedIn · Mastodon

Back to top ▲