You're designing an A/B test to evaluate the impact of showing content from non-friends in users' feeds. How would you test this with proper randomization?

To A/B test the impact of showing content from non-friends, we would randomly assign users into two groups: a control group that sees the current feed experience (e.g., primarily friends' content) and a treatment group that is exposed to a feed including content from non-friends (e.g., creators, public figures). The randomization would ensure that, on average, both groups are similar in all other aspects, allowing us to attribute any observed differences in engagement metrics (like time spent, likes, shares, comments) directly to the inclusion of non-friend content. We would define clear success metrics beforehand, such as increased overall engagement or content discovery, and monitor these metrics over a statistically significant period to determine the treatment's impact.