In the WCMS world, the big migration cost is in developing presentation templates. Most WCMS have a built in presentation tier with its own tag library or syntax. When migrating, developers need to re-build the presentation on the new platform from the ground up. This, in itself is a lot of work. Even worse is when the business takes the opportunity to lob in a bunch of “while your in there” template changes. Before you know it, you have stumbled into an un-managed web redesign project that you didn’t sign up for. Without some serious product management discipline, things begin to unravel very quickly after that.

One aspect of this is ensuring that media that is exported can contain as much metadata the customer wants from the host system. Either embedded in the native format, with XMP, or as an associate stream. At least this allows some of the context to carry with the media itself. So that it can reestablish itself in a new home.

This metadata rich approach allows for the data to be point of integration and interoperability rather than the aging system.

I think the strategy of the day was probably about keeping up with a changing marketplace and acquiring product categories that seemed hot. Vignette and Interwoven both felt that web publishing was running out of growth potential. They wanted to get into enterprise content management because selling a 100,000 seats for an intranet is a more lucrative deal that selling 40 seats to an internet publishing group. The customers paid the price because these acquisitions shifted these companies away from their core competency and the web publishing products stagnated. This is why Vignette and Interwoven both entirely missed web 2.0 and felt like an anchor to their customers who were competing against agile start-ups with more nimble technology.

However, in varying degrees, most of these acquisitions failed and the original application had to be scrapped and rewritten (ex. Bulldog was totally scrapped and DCTM’s DAM was rebuilt by the EMC staff up in Toronto). Worse still, a few companies kept the products and just built these elaborate integrations with the legacy products.

In some cases, the companies didn’t even attempt to integrate the solutions. When OpenText acquired Artesia, for instance, they kept that group as a standalone unit – I don’t know that there was ever any attempt to integrate that product into the larger OT stack. You can be sure that OT did push their premium products downstream to the Artesia group though.

I don’t necessarily agree that a single vendor couldn’t manage an enterprise content management solution. However, I wholeheartedly DO agree that trying to integrate completely different code bases and different company resources and cultures is a recipe for a disaster. I’m not sure anyone has ever been successful doing it in our marketplace.

Good post. -JB

This is why you should never use a CMS solution that is maintained and developed by a single vendor.

I’m sure you’d agree that there are other shades of grey: acquisitions where it’s more about acquiring complementary technology with the (sometimes misplaced) promise of tighter integration, or because the product lines had some correlation in the first place. However, I think you’re absolutely right that this is seldom the only or even main driver for acquisitions, which after all are usually about maximising shareholder value, not altruism towards the customers of either the acquirer or the acquisition.

Martin