I’ve got a CMS which a few of my client (marketing companies and other web development agencies) are using to allow their clients to manage their sites. Currently, the CMS is fully branded for each client, and they host it themselves with me setting it up.
Over the last year of so, I’ve found it increasingly difficult to manage off the different instances of the CMS. When I find a bug, I have to FTP into numerous different sites and upload new files, making sure it doesn’t break any customisation on that CMS.
As a result and as part of general improvements, I’ve development a new CMS which I plan to be sold as a package, with little customisation. My question is, I’m not sure whether I should offer this as a CMS that I will host, or the client to host.
The benefits in me hosting the CMS are that most of the framework remains hidden from the client (preventing copying/stealing) as the main files will be in a folder on the server. Also, because all instances of the CMS will use some of the same files, changes to them will only need to occur in one place.
The downside is that I’ll be responsible for the hosting, ensuring it’s kept up and running and the management of the server (although it would be a managed server). Some clients also like to feel like they’re in control of their clients sites.
Has anyone been in this position before? What would you advise?
At the minute, with the current CMS, I feel like I’ve got 10 different versions of it out there. With the new one, I want them to essential all be the same, but with a different stylesheet and logo.
I a big fan of hosting for the client exactly for the reasons you describe, or to manage a VPS or server for the client, which gives me the same control
Your processes can be improved greatly by using subversion or something similar to process updates on the clients sites. Of course this doesn’t work on shared hosting space
We offer this service with major open source CMS’s and frameworks (Joomla, Drupal, Symfony), because your self built CMS cannot keep up with security issues like these mass used ones can
We specifically offer the source code to the client, after all we want them to still use our solution if we are unable to service them.
Thanks for the reply, some hopeful information there.
The main reason I’ve developed a CMS was because my clients were finding systems like Joomla were overkill in terms of features for many of their clients. It’s also helped my clients gain clients of their own because they feel like they’re getting a bespoke CMS rather than an off the shelf product.
I’ve also found that many of my clients want the CMS to fit the clients site, rather than having to consider the CMS during the planning and design of the CMS.
I get these objections a lot as well, but there is a misunderstanding on how to use a major open source CMS ( I believe just because some [say Joomla] are designed to be used by non-technical people )
What you describe as overkill is actually irrelevant, because code and feature size do not impact usability, speed or anything in a reasonably designed system. People don’t call MS Word overkill, yet they rarely use more than 5% of its features.
Most installations are customised with usability plugins we write ourselves - because it is cheaper to write them than to find them. We are also not cheap and only use custom designs (with custom HTML), the client can match this up with receiving value.
There are 100s of ways how something like Joomla can be made fit to a clients site most of the times. The key is to not let the client go around and figure out which 5% functionality they need to use, but to tell them exactly how to use it and to ignore the other 95% of functionality. Business users do this with every other software package, why not with their CMS?
People also equate regular security updates with low quality software sometimes. The exact opposite is true. The software which has no found vulnerabilities contains all these issues as well - they just haven’t been found yet, usually due to lack of eyeballs.