In my humble opinion, I think that each consumer of Slack and other web services should implement their own monitoring. It probably would not hurt to open source the monitoring so that Slack and the ilk can help keep your monitoring accurate and honest. This would empower others to replicate your methods for the service providers they consume as well without having to pay monitoring service providers. They will only show outages from their perspective and some of them run from Amazon, which may be sub-optimal in this case.

Some existing open source tooling are Nagios and Sensu. Sensu being a little more dynamic and cloud friendly and a newer kid at the table. There are chef and ansible scripts for both of these.