Managed hosting / dev-ops service offering?

Currently the service proposition from is:

My hypothesis from my writeup of thoughts was:

My main hypothesis is that Webarch could do well by going for more of a “managed hosting” offer where above market price VM pricing aren’t a significant factor, and clients would enjoy the personal tech service that is already offered. So in this model the main offering is peace of mind that when the devs hand over the code, the rest is taken care of.

I asked my friend at what their experiences have been over the years and got this helpful feedback:

  • the market has significantly moved onto dev/ops consultancy - not just offering hosting services
  • they do a lot of external cloud management now (aws)
  • basic hosting/server management is not enough, devs can setup some basic aws/linode/digitalocean stuff, what they can’t do so well is CI/CD pipelines, complex hosting, etc
  • sales pipeline drives things forward, made significant progress after getting an experienced salesperson onboard - it’s about finding out what the client is up to and what projects they have coming up, what their needs might be, etc… different needs for different places
  • bare VM prices are higher than Webarch’s and monthly management fees on top of that (for OS updates, some app seutp, and troubleshooting, but not much more)
  • “setup” fee is now more a “initial build” professional services step, as working out what they need, building pipelines etc is quite involved

I think Webarch could be well placed to offer more of these complex/dev/ops/pipeline kind of projects from a technical level - as it’s the same tech as used inside the org (GitLab+CI, etc…) and you already do it to at least some extent (e.g. Development | Webarchitects and Technical Support | Webarchitects).

I would be interested to try and find out from some of the existing co-op clients what kind of help they need, and personally think Webarch is in a good position to build up a lot of decent dev/ops wisdom that the clients would benefit from (as they will have fewer projects to learn from).

My general feeling is perhaps more about becoming a partner for the clients that solve their problems, rather than a big list of products/services they can buy.

Next steps could be any/all of:

  • get some more discussion going on this thread
  • chat to some of the existing co-ops or other interesting clients to find out what they need in this direction
  • update website to add managed hosting, and dev/ops buzzword offerings
1 Like

I think that is all spot on @nicksellen :+1:

A couple of old examples of devops type work we have done is the work we did for the Transition Network which was all organised and documented on a Trac site (which we setup almost 8 years ago) and also the work for CRIN which we also used to use a Trac site for (setup almost 4 years ago). Both the Trac sites are now static archives and are not fully functional, the Transition Network migrated to WordPress and a managed hosting service that was cheaper some years ago.

CRIN are members and we have a new version of their webserver coded in Ansible on and this is due to be made live on one of our VMs at some point this month. Their current site is hosted on a pair of VMs we host, their site is behind Cloudflare.

Another example is The News Coop, they are also members and have a live and a dev server with us (behind Cloudflare) and again the code is on

For both CRIN and The News Coop they have other people doing the development work on their Drupal and WordPress sites and we have been doing the sysadmin / hosting side of things, Nginx / Apache / php-fpm / MySQL / Memcache etc configuration.

A third example is the Schumacher College, also members, we have configured a new, third party, physical server for them, the code is on (but private, unlike the others) and this has also involved scripting Moodle upgrades using Ansible as well as Drupal and WordPress hosting.

Sounds like some nice case studies to write up for the website too :slight_smile:

1 Like