Differentiate between when a hovercard is loading and when it is loaded/displayed. Also, all events are interactions so they should count as such.
At the moment, we are getting so many false positives in analytics about when people are actually looking at the hovercard. A user can accidentally trigger the 'hovercards displayed' event but just scrolling. From a product perspective, the timing is useful, but in its current configuration it's also concealing insights in analytics.
On #13, @saiichihashimoto said "If the timings of the hovercards are mostly low, that shows that they don't have much reason to keep the hovercards around (uninteresting content, not loaded, etc)". It is very difficult to differentiate the timing of the event from the event occurring, and still have useful data. Separating them into two events is better. That way, if the time of the 'hovercard displayed' event is low that means that once it loaded they didnt care. When both events are combined the timing doesnt differentiate between long loading times and uninteresting content.
Also the percentage of 'hovercards loading' to 'hovercards displayed' events shows us the percentage of abandoned hovercards.