More Features, More Headaches
Shipping Vs. Sinking: Why Adding More Features Produces Throbbing Pain in Your Organization and Product
In the age of AI everybody talks about how we're getting faster at adding. We will ship more features quicker. It's inevitable.
All this will do is produce a ticking time bomb that will make our existing problems worse. Unless we also get better at removing and editing all we will do is produce more bloat that drags everything else down.
More code is more headaches. More features are more problems. The hardest part about building valuable products is not adding something new.
Any fool can do that.
World-class products shed what is unnecessary to uncover the essential. Great product management is about subtracting. It’s about removing obstacles to clarity and utility from a product until the essential speaks the loudest.
More features isn't better. Less features that do the right thing are better.
Figuring out the right thing is still the hardest thing about building products, not shipping more things. Twice the work in half the time is twice the fluff in half the time.
What should we worry about instead of adding more feature? It’s pretty simple: how do we get better at removing?
Adding is the easy part. What should we remove to make something better is the hard part.
Removing is the difference between a mediocre and a great product.
In the immortal words of Dieter Rams: less but better.
Writing something is easy. Editing so it becomes great is the hard part.
Shipping more, means sinking faster and dragging ourselves down, unless we kill features at the same time.
I so agree with you, and wish more would think as you do. Back in 2009/10 I worked with a 25-year-old company that made industrial printers, both the hardware and software, and was falling behind new competitors. On exploring what was holding them back they identified "too many features" as the problem. The maintenance of specialised, barely-used, or even never-used features seriously hampered their ability to deliver actual value. Over the years I have encountered more of this problem. As you say, any fool can add new features. It takes skill and thought to add actual value, which often (as in the case mentioned) include removing all the years-accumulated bloat.