All of us have been involved at one point or another in our careers with that “death project” that just seems to lack any real conclusion and no one seems to know how or why it is in the state of limbo.  Vendors who serve the publishing industry have many reasons why a project is in jeopardy or has failed altogether including lack of proper resources, no project management discipline,  etc.   From a vendor’s perspective there are some telltale signs that were evident from the very beginning of the project but everyone overlooked them in the excitement of project kickoff.  Following are my top 5 reasons (from a vendor perspective) on why CMS projects fail at publishers:

1. Solution/Technology was not the right fit – Almost no one will admit to selecting the wrong solution or technology.  We all know that buying technology is sometimes a mystery.  Some vendors are really good at selling a vision only to have the publisher realize in the middle of the project that reaching that vision is going to cost three times more than they budgeted.  In other cases publishers already have a preferred technology or product and force that product on all groups.  On more than one occasion we were told point blank by a larger publisher that we love your RSuite technology but corporate is forcing us to use Documentum because we bought a site license.  Economically that makes sense. But functionally this may be trying to shoehorn a technology that was built to manage documents into a publishing environment that requires content management.  Different requirements, different solution required.  It is that simple.  It is no surpise when we hear back from the publisher 9 months later that the project failed or the system is not being used because the system does not meet expectations.
2. Buying a vision that is unattainable – Publishers get excited by vendor demos.  And they should!  What they are seeing in a demo is something they generally don’t have in place.  Some vendors are outstanding at selling a vision by demonstrating slick end-user applications.  The problem with this is that a publisher needs to ask the question “how much is it going to cost me to reach that vision?”  Seeing a technology vendor show really cool functionality does not mean there is a good business or production model behind it.  It is good to see demonstrations that push the envelope, but understand what it will take to implement such a vision (time, money, and business model changes).  I have seen several publishers purchase certain technology to build really cool end user applications only to have the technology sitting around because the vision was not attainable to begin with because it just cost way more than they could ever budget.  Investing in a vision is fine, just be able to break that vision down into logical, cost-effective projects.  Be realistic about what you would like to accomplish and what you can actually accomplish.
3. Poor project budgeting – Along with vague requirements goes poor budgeting.  If you went to a home builder and said "I want a two story house built, give me a quote." How much confidence would you have in the quote you would receive back when the requirements for the house were so vague?  Well, from a vendor’s perspective, we get this level of vague requirements for a CMS on a regular basis and are expected to provide a budget to implement the software.  It is generally couched with “we are only looking for a ballpark.”  OK, great, but if you are looking for a ballpark price that you would have a low level of confidence in, why are you putting that ballpark price into your next fiscal years’ budget?  Immediately you are putting the project at risk.  Again, vague requirements will lead to ballpark estimates that can be misconstrued in budgets. There can be pressure on the vendor to implement a solution based on an unrealistic budget.  Because the system does not operate according to some vague vision there is a real risk of project failure and unhappy customers.  See the chain of events?
4. Inherent conflict between IT and editorial  – My colleague, Lisa Bos, wrote several years back in one of our website newsletters that software development and editorial processes are in direct conflict with one another.  Think about it.  Software developers are used to an environment where they work up to the very last minute making changes on the fly and moving the system to production with an acceptable level of bugs.  The software is never 100%, but it is operational.  On the other hand, the editorial team has a defined process to complete edits on a deadline with the goal of 100% accuracy.  This inherent conflict between these operational approaches comes out during a CMS project implementation.  Understanding the cultural differences between the two organizations is important.
5. No definition of CMS project success – Why do publishers implement CMSs?  There are many reasons of course, but how often are the goals of the CMS project discussed: during the budget cycle, RFP stage, kickoff only, never?  If you hold a CMS project kickoff meeting and ask the group the definition of success when the system is operational and no one in the room knows the answer, you have a problem.  How can a CMS project be successful if the project team does not know the measurement of success?  Installing a CMS is not a success criteria.  Managing XML better is not a measurable goal.   Re-using X% of content in new derivative products, or reducing time-to-market by X days are real, measurable success criteria.  One exercise I like to do at the project kickoff meeting is to make the team draft a press release announcing the completion of the project.  After the team gets over the silliness of the initial request, most teams have fun with the exercise and actually contribute to the writing of the press release.  This simple exercise allows the team to verbally communicate among peers what their interpretation of success is for the project.  If you cannot articulate the definition of project success for the CMS project from the outset, you may be in trouble of ever meeting expectations of management.  Know the success criteria, communicate the success criteria, and celebrate the success with your team.

There are many reasons that CMS projects fail, but over the past decade these five are top of mind.  You will not be able to avoid all of them, but recognizing an issue early on and addressing it will benefit you in the long run and make everyone happier because of the ultimate success your team will achieve.