Since I was at university I’ve been working with and developing content management systems. For my final project at uni, I researched into the usability and accessibility of open source systems, most of which didn’t rate highly.
At work, we’ve got our own CMS, which was rated highly in usability tests by a client although it’s far from perfect. I think it’s easy to be critical of things you’re close to.
It’s about due for a re-write. We’ve had many versions which have had (sometimes) large amounts of customisation, dependent on the needs of the client. Through this, we’ve discovered a lot of what clients want, what the system needs to do to better accommodate these and more importantly, often how little the client uses some of the custom features they have requested.
Like many developers, I’m a fan of 37signals products; they’re simple (to use) and have the principle of developing products for themselves, which not everyone agrees with. For the products they offer, which help with project management and collaboration, that kind of makes sense; they started out like many web agencies doing client work and so made products that would help themselves, which they thought other people could benefit from. This has pretty much made their business, they now have a great income from these products and a great business.
I’m wondering if their model could work for a CMS. Would people/companies use a subscription based CMS product? As many CMS are so overloaded with features, it can be difficult to distill down to a core set of features what makes a good CMS. The market for CMSs is also pretty swamped with so many small to large open source projects and cheap to expensive off-the-shelf system as well as custom systems used by development houses.
Can we design a CMS for ourselves as technically minded folk or can we rely on techniques like personas to make a simple usable product? Is there a market for another CMS?