Metrics:

Number of times the Watchlist Feature is opened per week
Number of articles viewed per session
Number of articles saved to the watchlist
Number of Watchlist Feature who return to the app within 7 days
Number of users who return to the Watchlist Feature within 30 days
Number of users who use the Watchlist Feature at least 3 times in a month

Duration: The experiment will run for 4 weeks.

Experiment Plan:

Randomly select 50% of the target audience to be in the experiment group and give them access to the watchlist feature.
Leave the other 50% of the target audience in the control group without access to the watchlist feature.
Monitor the metrics listed above for both groups for 4 weeks.
Analyze the data collected and compare the metrics for the experiment group to the control group.
Use statistical methods to determine if the results are statistically significant.
If the results show a statistically significant increase in user engagement and retention for the experiment group, roll out the watchlist feature to all users. If not, re-evaluate the feature and experiment with other ways to increase user engagement and retention.

What does engage mean?
Newcomer <50 edits
Medium 50-1000 edits
Expert >1000 edits

(1-failure, 2-somewhat interesting, 7-good product, 14-great product days)
30- part of digital life

How do we classify frequent vs infrequent users?
Add to watchlist
Update expiry
Remove from watchlist
View watchlist homepage
Go to diffs from homepage
Click undo, revert, thank
Mess with filters