Paul's Internet Landfill/ 2017/ Dealing With No-Shows

Dealing With No-Shows

When organizing activities, dealing with no-shows (people who promise to attend your activity and then don't show up) is a huge problem. If you are trying to keep the event accessible by not charging for it (or charging a trivial amount to cover some expenses) then the problem gets worse.

One solution to dealing with this is to make your event expensive, but this will reduce turnout (and sometimes reduce turnout among the very people you would like to attend the event).

I have been thinking about how to deal with this problem for a while, and I have two proposals for solutions.

Charge People For Every No-Show

Say you are running a class. You can charge a large up-front fee for the class, and then refund that fee to everybody who shows up. The no-shows do not get their money back. If your class has multiple sessions, refund a portion of the fee for each session.

In principle this keeps the event accessible, because people end up paying nothing if they show up. In practice this shuts out poor people, who do not have a lot of capital to gamble on your class. There may be some workarounds for this, but the ones I have found incentivize bad behaviour by the poor.

The fee has to be big enough that no-shows will feel the impact of not attending. Given income disparities in this country, finding an appropriate fee becomes difficult.

Despite these flaws, I would consider a model like this if I was running a class for something like KW Freeskool.

Assign Trust to RSVPs

I am under some pressure to organize the Spectrum Outdoor Recreation Group, but I won't do it.

If I could depend on RSVPs then organizing these recreational activities might feel okay, but the thought of organizing an event and then being the only one showing up does not appeal to me. Too bad RSVPs are useless, whether they are Facebook event signups or Meetup RSVPs.

Here is an idea: for each event, assign a minimum number of "RSVP units" which would make the event feel worthwhile. Publicize the number of RSVP units necessary in advance. If you do not receive enough RSVP units to carry out the event, cancel it.

Now classify everybody who RSVPs to your event into one of three states:

Why would an untrusted person RSVP at all? To improve hir status, because if there are not sufficient RSVP units to run an event, then the event won't run, and presumably some people would like to attend the event.

If you have to reduce your membership (perhaps because Meetup charges more if you have over 50 members) then dropping untrusted members is a good strategy.

I do not know whether these thresholds are correct. Maybe first-time RSVPers should be semi-trusted and not fully trusted. Keeping track of these logistics could get messy. But I think the basic idea is sound, in the sense that it gives you a principled way of evaluating whose RSVPs you should trust and whose you should ignore.

This model only works for events that are held fairly frequently.

The math is not elegant for the numbers I chose. If you change the model so that semi-trusted people are worth 1/3 of an RSVP unit and new RSVPers are semi-trusted, then the number of RSVP units should be approximately the number of attendees you expect.

How many RSVP units should be required to run your event? It depends:

What should happen when somebody RSVPs, and then unRSVPs at the last minute? Is that a no-show? This policy can get complicated. I think the number of rules you set up depends on the size of your group, and how much you trust it.

I like that this strategy does not punish people who don't RSVP and don't show up. I also like that sometimes more people will show up than you anticipate (which is usually a nice problem to have), because some people won't RSVP and some semi-trusted people will show up. But this is just a cutesy idea in my head; I have never tried it out.