Always ask – What is the people problem we are trying to solve? (edited)
So by changing the setting from “Friends” to “Friends and Friends of Friends” – are we hoping to let people share their posts more broadly and there by get more likes, comments, reactions, shares etc and create a feeling of higher social validation? Will this create a virtuous cycle of more shares/posts/comments etc – i.e- more content creation and more user engagement over all?
Are we hoping to also see an increase in number of friends/connections per user – as posts will be shared with friends of friends who will then reach out and send friend requests – higher activity within distant social graphs?
So we would do an A/B test of people (control group) who will see only “Friends” and test group who will see “Friends of Friends”
See if you have higher engagement rate within the test vs control group by testing for total and average number of shares, likes, comments, reactions across the two groups
See if you find more friends being added in test vs control group for the same period of time (30 day period might work)
If the engagement and number of friends metric goes up – this is working
On the other hand, if this causes, anxiety and fear of privacy issues, you might see a reverse impact and user engagement may go down and number of friends may also go down. Then- this feature will automatically be deemed as a bad idea
We should also allow a user to change it back to “friends only” and measure if in the test group, people actually use this toggle to go back to original setting. This would show if we are getting adoption of this feature
What I mean is – if people are self selecting out of this setting consistently, its a very clear sign that people are not comfortable with this setting.