4 ms·

Thanks Ragnar :) There seems to be consensus that per-call is the way to go, going to work on this soon.

• aristus 18y ago
  This is awesome. I was going to build something myself with ribbit. Think over the pay-per-call idea very carefully. On one hand it feels good to potential customers, but if they are in a real emergency, knowing that each of those little robo calls costs them 1 euro or whatever will start to grate. Also, pay-per-call diverges the incentives. They are essentially making a bet with you. The better their uptime (good for them), the less revenue you get (bad for you). It's dangerous to build in a divergence like that if you want repeat business. You might want to charge a yearly fee plus a small charge per call.
  • mmelin 18y ago
    Thanks! :) There are a few API-type options like Ribbit available for those who like to tinker and we actually considered using one of them first for RoboNOC. However the value-add we're trying to provide is both ease of use but also redundancy and security, and so we built our own infrastructure to not be reliable on any one third party. You have a very good point on the pay-per-call. Going to pay-per-call will of course drastically change the revenue models and forecast, but is of course (or at least looks like) a better proposition for new customers. Combining per-call fees with a yearly charge might be the best of two worlds but will also need careful work to communicate effectively and not complicate things.