+3
In Progress

Anti-bot protocol

Curtis Stensland 3 days ago in Reservations updated by Cali Slaughter 2 days ago 1

I know you are aware of the issue of members booking courts using bots. We have been dealing with this as a club for about 4 years and have a LOT of experience with all the tactics and tricks members will use to take advantage of the systems in place. I know you have released a feature that will allow for one timeslot only per day to have a reservation lottery. Unfortunately this is not very useful to use as we have members booking using bots at all times of day.

There is another tactic used by bots that I like to call "walking down" where a person wants to play at 7pm-10 but there is a lot of competition to book the courts, or perhaps there are other bots they are competing with. This person will book a court from 9am-12pm and right before 11am will cancel the booking and immediately book for 11am-2pm. At 1pm they will cancel and book from 1pm-4pm... and so on until they get to 7pm. So, someone waiting to book at 7pm will never get a chance to book the court since this bot was walking this reservation down the scheduler. We are wondering if a couple suggestions can be implemented that would help level the playing field between humans and bots (at least for now).

1) captcha. Would adding a captcha prompt during the booking process help prevent the usage of bots? Could this be an option toggled on/off in the settings so it can be optional... perhaps some clubs like the bots. I will say bots have really helped some groups book courts efficiently so maybe not every club wants them gone.

2) Flagging sketchy behavior. This would be my preferred method of handling bots by Court Reserve recognizing when an account is engaged in patterns of booking behavior that are either impossible for a human to do, like instantly booking multiple reservations, or walk down a reservation by repeated cancelations and re-bookings. I'm thinking after an account gets flagged Court Reserve can send them a warning and give a notification to admins that an account has been flagged. If the behavior continues Court Reserve, with permission of the admin, can remove booking privileges for this member. We have a feature for penalty cancellations, but not one for penalty behavior. There could be a counter and this sort of thing could be automated to some degree for the admins. 

That's all I have for now. My staff is very eager to fight back against the bots. We have quite literally thousands of software engineers that are members of our club so we run into a lot of these issues. They are very clever and capable folks, so putting a stop to this behavior is a good challenge. Please let me know how I can be of any help. Thank you.

In Progress

Hi Curtis,

Thank you for your submission. The Product Team is actively working on a solution for this. The initial implementation will include:

  • automatically flagging potential bot activity for reservations (based on unusually high volume of reservation attempts, rapid-fire submissions, and multiple actions in <1 second)
  • ability to implement member-level or global-level CAPTCHA for reservations
  • outlining reservation activity/potential bot activity at a member-level and in a global bot detection audit log

Admins will be responsible for determining how best to proceed, such as reaching out to the member to let them know that bot activity has been detected, implementing CAPTCHA for members based on findings, etc. I will make sure to document your feedback, to be considered in potential future iterations.

The Product Team will provide updates as this progresses. We anticipate conducting a beta period for this functionality soon.

Thank you again for taking the time to submit this idea.

Thank you,

Cali