I assembled a list of Discourse hosting providers to help a client with their hosting decision. As we talked through the options, I started to realize none of them were ideal:
- DIscourse Basic is the cheapest option on Discourse Hosting. A quick look at what it provides, 50k pageviews in particular, gives the impression it’s ideal for building a small community. But a closer look shows the plan is limited to 100 member who must be invited in order to see anything on the site. I’m honestly not sure who this is for. (If you want to try out Discourse for your club or team, consider creating a group here for $0.)
- Discourse Standard costs twice as much.This isn’t really a problem for established forums, but it’s a lot to ask of someone starting a new community.
- Communiteq Starter at $20 is more like it. The problem for most of my clients is that plan doesn’t include the Discourse Subscriptions plugin, so you can’t fund the forum using a subscription model.You can run ads, but that’s not normally a productive way to build a new community.
- Communiteq Professional can include Subscriptions, at a dollar less than the Discourse Basic plan. It’s the option that seems most appropriate for my clients, but still substantially more expensive than what I pay for my own sites.
That’s when I hit upon the idea of offering my clients hosting that I manage. I can charge a smaller fee by using my understanding of Discourse to host on DigitalOcean or Lightsail. My clients are happy because I lower their costs and I get a steady revenue stream. If my clients’ communities grow past my ability to support their hosting needs, the other options make more sense anyway. The more I thought about it, the more I liked the idea.
One benefit for me is that the money I save clients on hosting can be used to take on initiatives that will help the client’s community. For brand new communities, that might be finding ways to advertise to potential community members. Or it could mean hiring me to implement a feature on Discourse or correct cultural problems in their community. Or my clients could just pocket the savings to stay viable as a business for years to come.
To be clear, I don’t necessarily want to be in the Discourse hosting market. If I thought one of the other services would work for my clients, I’d much rather let other people take care of that need. I don’t have to be the one to fill the gap. But as long as the gap exists, I’m going to do what I can to help my clients succeed.
A note on Service Level Agreements: I’m copying the Communiteq’s 99.9% availability. This is sometimes called three nines and means I can afford 8.77 hours of downtime a year. If I wanted to, I could plan a weekly 10 minute maintenance window and still have time left over. Despite doing some work that required downtime this month, this Discourse site is well within that standard.
In my experience, downtime is unpredictable and liable to last longer than you’d expect. The good news is that it’s relatively easy to spin up a new server for testing and most of my clients have small backups. I wouldn’t want to commit to hosting College Confidential which takes many hours to restore from backup and has millions of pageviews every month.