Friday, September 22, 2006

Picking a CMS

We've been very busy, trying to finalize a decision on a new CMS. After going through the nightmare of developing requirements, publishing an RFP, dealing with sales guys and doing demo after demo - we finally reduced the vendors and choices down to our finalists.

I've bought and implemented several CMS solutions over the years - Interwoven, RedDot and several custom built ones have taught me a lot. I thought it would be interesting to share the core selection criteria we used given how different this was from my past experiences.

At a strategic level, we had 2 main requirements that needed fulfilling before any vendor could possibly have a chance at winning our business:

  1. Integrated Web Services (direct control and implementation of web services at the template level)
  2. Very sophisticated multi-affiliate/micro site management
What's interesting about these 2 requirements is that:
  • Most all vendors say they can do these things (but most can't, or don't really even understand the requirement at all)
Instead of continuing the story on to who we picked and what we put them through, I thought that I'd just simply explain our 2 requirements a bit further so that you understand why these things are so important to us at this time.

Integrated Web Services
Web services as a term is a slippery one - it can devolve quickly into a catchphrase like "Web 2.0" and become meaningless very quickly. However, we envision web services as a layer that sits between a point and click interface, and an extremely complex back end system (home grown for now). When I collaborated & fantasized with our IT team about what the ideal solution would look like, we often said that marketing shouldn't have to ever call IT to build or customize our web sites. Everything should be completely customizable, including data access to and from our back office systems.

By having integrated web services built right into the templates in our CMS, we can actually begin to build interfaces so that the marketing team can develop custom applications on their own - no IT needed (this may still prove to be a fantasy, but one that is worth pursuing). For example - a calendar widget that we can 100% stylize and customize but that has hooks into our back end scheduling system. That way, we can centralize event data and dates - but repurpose that content anywhere we want.

In addition to never needing IT, we also want to make giant strides in back office workflow. Our current/legacy systems all treat the web as a completely different channel. Nothing is integrated, data never flows freely between our systems. For example, if you get our email newsletter (sign up here!) and change your home address via our member center, that data never gets back to our campaign or DM staff. Integrated web services solves this by creating a centralized data store for our data along with an easy way to integrate external applications - in real time - no old school, unreliable feeds required.

Trick is... hardly any CMS vendors we looked at have this concept baked into their systems. Oh, they'll argue with you, but as soon as you look under the covers you'll certainly find what we found :)

Very sophisticated multi-affiliate/micro site management
The 2nd requirement which centers around our chapter sites is fairly cut and dry at one level - provide a way to create and clone sub sites that are tied to a main site. No real trick there...

The twist, borne out of years of having to deal with over zealous chapters creating pages with pink backgrounds and light green text came about when we were talking about what it would really take to control our brand online.

The conclusion we arrived at was that we needed a set of tools that provide us the ability to lock and load content at the field level - not the page level. For example - on today's LLS site chapters cannot edit the "L" frame that houses the branding, main navigation and footer. On any given page however, chapters may alter the body of the page (view example). Within a given page, chapters can add colors, mis-use photos by loading huge, uncompressed images and pretty much do whatever suits them.

The solution I believe is in how you control content at the field level - and that is where most CMS vendors fall on their faces. I want to be able to lock down via CSS or simple controls everything that has to do with branding and design - leaving chapters to do what they do best which is provide local content and information.

While there are a few point solutions out there that satisfy both of these requirements (Ektron for example), no CMS vendor we've looked at quite has them both right.

The vendor we'll probably end up with has nailed the web services, and is promising future enhancements around field level controls. I'm not quite ready to publish the name of the final vendor, but will do so eventually!

Finally, a quick word about open source CMS solutions (yes, we did look at Joomla, Plone and Drupal) - but determined very quickly that for our scale these systems were not ready for us primarily because of lack of workflow. That said, if we didn't have to worry about 66 chapters and multiple sites - Drupal probably would have been a very strong contender.

Tags: , , ,

Links to this post:

Create a Link

<< Home