Erring on the Upside

Becoming a convert to Waze over the summer, I’ve started to use it all the time, even on routes I use everyday, just so I can be informed if there are any unusual problems. I like the mild gamification and that there’s a way to see who’s around you. Also, it’s great to know when and where a traffic jam will end.
One of the things that I’m impressed with the most is a simple implementation. For what I’ve saved for “Work” or “Home”, it predicts that I’m heading towards there based on my previous trips and will automatically load the directions as soon as I start the app.
Sometimes, the app is wrong… I’m not heading to a particular location at a specific time. However, it’s possible to cancel and enter the new location or address fairly easily.
When I am heading to a location and Waze preselects it, it’s magical. It seems like the future. The experience is many times more happiness inducing than the false directions are annoying.
I’m wondering where erring on the upside might applied in other places of product development.