It is worth writing down what a year of running programs actually taught us, including the parts that did not go well. Annual reviews that only list wins are marketing rather than learning.
What worked
Library partnerships were the single best decision we made. Hands-on curriculum outperformed presentation-based material by a wide margin. Keeping workshops free preserved the audience mix we wanted. Writing publicly forced clearer thinking than internal documents ever did.
What did not
We over-scheduled early sessions and had to cut material live, which is a poor experience for everyone. Our first evaluation approach measured enthusiasm rather than learning. We underestimated how much of the work is logistics, and volunteers who signed up for science ended up doing setup.
We also promised faster progress on Rise2Research than a student team could deliver alongside school. Timelines built on optimism are a recurring mistake.
What changes next
Shorter, better-paced sessions with a firm content limit. At least one evaluation measure capable of returning bad news. Written runbooks so a session does not depend on any one person. Fewer commitments, described more precisely.
The general lesson is that constraint is not the enemy of ambition. Doing fewer things carefully is what made the year's work worth continuing.