If a user has paid for the time, he should be flawlessly taken into the conference if it is still running.
The idea is the user should get the best experience when s/he has pre-paid for your services. That increases the responsibility on your shoulders to the point where you might want to give some benefit of doubt to the user for disruptions.
I am reluctant to introduce another 15 mins counter. While the system sets up the call, the system should,
- Inform about the total time of the video conference.
- I would not add disruption related information upfront, assuming the platform is stable.
- In case the platform is not stable, users should be informed about the negative scenarios.
The errors can be on two sides, either the system messed up or the error is on the user's side. In either case, when there is a disruption and user attempts to open the app again, the app should flawlessly take the user into the ongoing conference. Here we are only discussing app crash. In the case of network failure or failure on the side of the consultant, a relevant information should be displayed to the user.
I only highlight the flawless reconnection because, in case of paid and timed events, users are anxious about the value for money they get. Disruptions and app crashes are taken very negatively. The system can proactively increase the stipulated VC time by offering an extra time and clearly informing the user that since there was a disconnection (because of system failure) an extra certain amount of time is provided as a compensation. This will go long way in building customer satisfaction. When the system acknowledges issues and tries to compensate/rectify right away, it is taken as a positive sign (purely experience, no data to back up this statement)
I am thinking about the way online exam servers work. If there is a failure or disconnection, the system attempts to reconnect when the app is launched again. The attempt is to minimise the stress of the user when unforeseen things happen.
If the user fails to connect back in the stipulated time of the conference, the system should notify the user with an Email and additionally, an in-app notification about the last conference status.
Furthermore, it is imperative that the crash/error is analysed. If is is system failure a compensation/refund or an opportunity to reconnect at a later suitable time should be offered.
Additionally, the system can take precautionary measures by measuring the signal strength and reliability before making the call. This can help notify and warn users that there is a chance of disruption and we may and may not be responsible (depending on the nature of risk). As the platform matures, the app crashes will reduce but the signal strength issues will continue to trouble. So there should be a permanent solution to address that sort of a thing.
Wrote a little too much! I hope that is helpful.