Scatter and Sweep

One of my favorite one-on-one topics is “learning”. I get lots of questions in one-on-ones that lead to this topic: How did you learn when you were an engineer? What blogs or books do you recommend? And a lot of my answers really go back to the core question of “how do you learn?” Regardless of the engineers’ favorite media for learning, there’s an approach I’ve found that always helps me both see more information and retain it for when I need it. I call the approach Scatter and Sweep. Before we go into how it works, let’s start with some first principles.

First Principles on Learning

When I was figuring out better ways for me to learn new material, I started by asking “why?” It seems like a funny question, but it led to a few pertinent answers.

Those are great principles on which to build. I’ll translate those to:

Now we have some great goals for our learning framework.

The Scatter and Sweep Method

The Scatter and Sweep method is meant to give you the chance to see a number of different problems and solutions to promote efficacy, repetition so that you can quickly recall problem and solution matching patterns, and depth so that you can strengthen your teaching.

Start by imagining everything you want to learn is written on index cards. On the front of each index card is the subject matter. You can imagine topics like Git on the Command Line, Giving Difficult Feedback or, hell, Exiting Vim. On the back of each card is everything you want to know about that subject.

When you Scatter and Sweep, you’re trying to learn all of this information in two ways. Scattering is researching and looking up information on the problem you’re currently trying to solve. Sweeping is starting at the first card, reading it, and moving on to the next one.


By scattering and sweeping, you’re able to double your efforts on learning. You’re also able to approach problems from two distinct paths which can help lead to increased knowledge retention. And it increases the chance that the first time you encounter a problem, you’ll have a playbook for it.

I think it’s common for a lot of learning to happen via the scattering approach. If you’re at a job, and you don’t know the best way to solve a problem, you’re likely putting in some research to get the job done. It might be less likely, however, that you are adding in the approach of sweeping.

Another common refrain I’ve heard is that folks will hit the books but never see those problems come up in their day-to-day. There is a key factor here. Your sweeping should be done close to the same surface area as your scattered learning. You are trying to increase the chances that there is overlap. Think of overlap as though you are applying multiple coats of paint. This is not unnecessary work. Instead, you are strengthening the knowledge you already have. You want to feel the thrill of seeing a problem for which you have recently read about a solution. This, in turn, will encourage you to continue sweeping.


Regardless of the subject matter or the medium for the sweeping, I’ve found the Scatter and Sweep method to be an incredibly effective framework for learning new things. Next time, I’ll talk through the practice of how I apply Scatter and Sweep to learning and improving on my skillset as an Engineering Manager.

Dan Ubilla is obsessed with the craft of engineering management

He writes every two weeks. Sign up below for early access to his blog posts.

    We won't send you spam. Unsubscribe at any time.