The Art of Simplification: Why We (over) Build When We Should Simplify
Simplifying can add as much (or more) value as building new things. The value is frequently overlooked, which creates an unfortunate trend of spending too little effort simplifying.
Welcome to the Scarlet Ink newsletter. I'm Dave Anderson, an ex-Amazon Tech Director and GM. Each week I write a newsletter article on tech industry careers, and specific leadership advice.
Free members can read some amount of each article, while paid members can read the full article. For some, part of the article is plenty! But if you'd like to read more, I'd love you to consider becoming a paid member! The original version of this article was written in 2021. I love the topic, so I’m re-writing it.
Creating things gets the majority of our attention and respect at work.
What do you view as the bulk activities of work? We think of things like launching new mobile apps, building new features, and documenting new processes. These are the activities we associate with valuable work. They are the types of activities core to our career growth. They’re at the center of promotions, raises, and press releases.
And I agree that creation is wonderful. Inspired creative effort created the iPhone, Stripe, and the Bar Raiser process at Amazon (which I view as one of the miracles of modern corporate process, in that it doesn’t suck).
While these products and processes were indeed new things when they were created, they were not only the result of creation. They were also the result of elegant simplification. The iPhone has a user interface famous for its simplicity when it launched. Stripe is known to be one of the easiest payment platforms to use by developers. The bar raiser process has been used for millions of hires, but you could document the entire process on one page.
Value creation via simplification.
Simplification is an underestimated form of value creation. You can remove complexity to make something easier to use for customers. In fact, many attribute the success of products like Stripe and Instagram to the removal of non-essential features. That simplification allowed them to be built quickly, but also allowed customers to easily understand the tool, adopt it, and happily share it with others.
Software products are rarely made faster by adding features or code. When you are trying to improve the performance of complex software, the most drastic impact comes when large portions of features or code are completely removed.
Early in my career at Amazon, I had a valuable learning moment. We had a critical short-term deadline once for a major project. We were contractually obligated to get it done by a specific date. As a side note, this is very unusual at a company like Amazon. In the 12 years I was there, this was the only time we had an external date commitment.
There were four engineers working on this project. This four-person team fell behind schedule due to some complexities (which we probably should have anticipated), and raised an alert. They said it was now impossible to hit the date.
We had a quick brainstorming meeting, and discussed options. Everyone agreed that adding people to the four-person team would slow them down. There wasn’t enough work to split to a fifth person, and coordinating with yet another engineer would slow things down.
We couldn’t cut scope, as the specific requirements had already been simplified as much as possible. The engineers on the team brainstormed, and said there was no reasonable path to completing all those requirements on time.
What did we end up doing? We took those four people off the project, and put one (excellent) engineer on it. That person hit the date as planned. It wasn’t easy, but they nailed it.
How is it remotely possible that one engineer could do what four couldn’t? Were they one of those mythical 10x engineers? Well, perhaps that’s part of it. But the real win was that the team working on the feature was drastically simplified. That engineer had zero process to follow. They didn’t need discussions over APIs, didn’t need to integrate with other people’s features, and were never bottlenecked on someone else’s work. In essence, we simplified the task to “build these 29 things”, and let them put on headphones and work.
You can’t usually operate within a large company without process and coordination. But it highlights a key insight. Every bit of process is a cost. Communication with other people is a cost. Coordination of work between people is a cost. Every single effort not directly creating the necessary value is a cost. And simplification is all about optimizing value.
In large companies, process and communication and coordination can take a significant portion of feature development time. You have leadership review time, and program review meetings, and team meetings, and email threads, and goal setting meetings, and checkpoint reviews, and metrics reviews. The list goes on. Reducing this overhead can be the biggest lever for an entire organization to move more quickly.
We focus heavily on building, and scaling. While some of our biggest gains come from simplifying. Not doing something. Not having something.
But the value of simplification is (frequently) not recognized.
Unfortunately, the focus on creation leads to simplification being undervalued by peers or management. In large part, it’s because creation is visible, and simplification is the absence of things.
When you hire a large team, your large organization is obvious. We can all see that you’re influential. Heck, last I checked, the number of people reporting to you was one of the top data points in any manager’s promotion document at Amazon.
When you are extremely efficient with a tiny team, what we see is a small team. It’s challenging to imagine the absence of the team members that you didn’t need because you were so efficient.
When you resolve a large issue for customers and ship it, the value is obvious. Everyone can see what you did, and the value you created.
When you avoid a problem for customers through careful removal of technical debt, the value is invisible. If you make it easier for customers to use your product by removing some buttons, the product becomes subtly more sticky. It’s challenging to acknowledge the resolution of a problem which never happened.
I’m going to walk through a time on one of my teams at Amazon where simplification clearly had a higher value than adding something.