One of the interesting conversations coming out of my last post about Product Guys is the need to cut features.

It seems like everyone in Silicon Valley likes to blog about how "a good product should cut features" when they have zero track record of actually doing so.

It's an easy thing to say. But it's really hard in practice, much harder than adding new features.

To cut a feature, you need to understand your user base deeply. You need to have metrics to back up your decision. You need to understand the impact.

You will lose some users. Some of your earliest and most loyal users may love that feature you just cut. And they will complain, loudly. They will tell the world that your company has lost its way, jumped the shark. It will hurt.

But that's why you have a good product guy. That person should understand the impact of cutting the feature. And they should have a clear reason why cutting this feature will be good for the product in the long run.

They should be able to tell the board that cutting that feature lost 1% of the user base, but tripled growth. They will have confidence in the decision.

The worst thing that can happen is to be surprised by the response of cutting that feature, and reverting the change. You will second guess yourself forever.

Adding features takes creativity. Cutting features takes balls.

You should follow me on Twitter here.