Test Small, Release Big

There is a reason that when you deploy software, you have alpha and beta testers.

Why?

Cause shit goes sideways when you mass deploy.

We don’t live or operate in a vacuum. So when you are building something new, you need to test with smaller sample sizes in the wild before pushing to entire groups. This helps you avoid pain, confusion, and overall negative user experiences.

Luckily the Central Limit theorem helps us solve this problem!

In short, the distribution of sample means approximates a normal distribution as the sample size gets larger, regardless of the population's distribution. Sample sizes equal to or greater than 30 are often considered sufficient for the CLT to hold.

Simplified, you will essentially have all of the data you need after getting results from 30 (random) people or so.

So, before you release anything out at scale, run some tests. At a minimum, run it past 30 people/customers. This goes for software releases, marketing campaigns, sales pitches/adjustments.

Very important to keep this in mind with everything that you do while building a company!