The Bifurcation of Content Management and Delivery

At the moment of publish, your primary concern about content changes You move from a primary concern of “management” to a primary concern of “enhancement.” Before publishing you were concerned about modeling content, creating it, controlling its workflow, etc. After publish you’re concerned with optimizing its delivery.

These are really two different disciplines, and they can’t be both lumped under the rubric of “management.” So many things are happening in the delivery tier now (post-publish), that you almost have two different systems:

1. A Web Content Management System (WCMS) – what happens before you publish

2. A Web Content Delivery System (WCDS) – what happens after you publish, when your content is “in the wild”

(Related: The Dawn of the Web Content Delivery System.)

What I find interesting about this bifurcation is that most new features are for the delivery tier, not the management tier. New CMS releases hype huge functionality changes that only come into play post-publish and that are all about “optimization,” not really “management.” Vendors tend to move in packs. A couple years ago, A/B testing was all the rage. In the last year, it seems to be personalization.

I think there are two reasons for this.

First, core content management is a somewhat solved problem. The actual management of content – the mundane, repository-level things like creating content, updating it, applying permissions to it, sending it through workflow, etc. have really been solved to a degree of acceptability in 99% of use cases. I’m certainly not saying there’s never room for improvement, but a lot of what we want to do in this area has been done.

And from a sales perspective, how do you make repositories more interesting? How do you make them do more than they do now? How do you make them better at letting you model, create, edit, and store content? When you’ve covered the big stuff, you reach a certain level of feature saturation.

Second, content delivery is…sexy. It sells well. Things you can do in the delivery tier like A/B testing, integrated analytics, visitor segmentation, etc. demo great in sales situations. People respond to these features, they see themselves using them, they visualize their own content in these features.

By contrast, if the high-point in your demo is, “Here’s how you enter content,” you have a real problem. All CMS can enter content. The delivery layer is where the market differentiation occurs. It’s the place where a vendor can do something that no one else has done and set themselves them apart.

Now, if we accept that this bifurcation has occurred, we come to a really interesting hypothetical –

I believe that, at some point, a vendor will decouple their delivery tier from their management tier and spin it off as a separate product that works with other vendor’s management systems.

Either that, or a vendor is going to come along with such a compelling, standalone delivery product that they’ll achieve a critical mass of CMS vendors who develop “connectors” for it. So much so that eventually customers will be able to pick a WCMS from one vendor, and a WCDS from another vendor, then hook them together, with the WCMS pumping content into the WCDS to be delivered.

Episerver CMS v6 R2 has one of the best features I’ve seen in a CMS in its Visitor Groups. This feature lets you develop criteria to profile demographic groups – you can put all people who come from monster.com, for instance, into a “Potential Job Seeker” group and then morph your site for them – highlight a different featured link, for instance, or even show different content down to the paragraph (even sentence) level.

The feature is extraordinarily well-planned and executed. I’ve demoed it and it’s really a category killer. When prospects see Visitor Groups in action, they suddenly stop talking about any other vendor. At least one prospect literally went into what can only be described as “slack-jawed amazement.”

This feature is delivery-centric. There are some things you do on the management side to enable it, but it doesn’t change how content is stored or fundamentally managed. All of the impact of Visitor Groups happens post-publish, after content has been “managed” and is now being “delivered.”

So, it’s got me thinking – how long before someone abstracts this feature? It happens to be part of Episerver now, but it could really be done with any CMS…or without one. How long before someone develops some CMS-agnostic system that runs in the delivery tier that provides this functionality? They’ll sell some connectors, of course, that make it work gracefully with the popular CMS, but it would stand on its own.

Alternately, what if Episerver ever decided to spin this feature off? I highly doubt they ever would, but what if they sat around and thought, “We could make more money by making this a standalone product and selling connectors to hook it up to Sitecore, Ektron, etc.”

So much is happening in the delivery tier right now that we’re going to hit a point where some vendor realizes they could do better selling a delivery tier system for any CMS, not just theirs. When this happens a CMS will slowly become a true content management system that powers a delivery tier that likely comes another company that sells nothing else.