Event Drop-Off by Month...

Partners spend a lot of time asking about effective execution of events and which variables have the most impact.

Oftentimes, I say that it is The Long Runway in terms of demand gen. That was anecdotal, now I have some data which supports (sort of) the hypothesis.

These are 8 partner-led events in Mid-Atlantic that we tracked in terms of Registration and Attendance by Month for April. So, in this case, Month 1 is March, Month 2 is April (I will be tracking going forward as well).

All of the events (except Partner 8) took place at the end of April.image

The list is sorted by those with the highest attendance. Partner 1 is an anomaly in many respects. You could also argue that we don't have enough data yet (which is why I will be tracking going forward).

Some observations...

  1. The drop-off of customers is not equal month to month. More drop-off the farther out you go.
  2. However, those who registered more than 30 customers more than 1 month in advance tracked towards a successful event (more than 25 people). As Partner 4 shows, it can be done, but your odds are better if you start earlier (at least that's how I read it).
  3. Those who started earlier became better event marketers with practice, with on average 23% less drop-off than those who were ineffective more than one month out.

What am I missing? What do you see in the numbers?