More and more companies are being faced with similar challenges now that most firms have made some type of attempt at rolling out a CRM initiative or customer database.  Gone are the days where customers are implementing from scratch so there is an entire level of discussion on how to deal with any existing systems and how to move forward with any new initiative.  We need to assume in most instances that there will be some "system of record" and then assess the solution based on its operational value today and not the time or investment made in deploying the solution initially.

History also tells us that it is usually not all or nothing when it comes to making decisions on the both the database and the data.  I guess what I am saying is that the decision to start from scratch or ehance/modify/re-implement will really be dictated by what the existing capabilities are and how they align with your needs going forward.  Too many people confuse the data in a database with the actual value of the system that houses it.

One of the most confusing elements of a project is dealing with the various data sources and tying the data into any new initiative. Understanding the technologies that exist today the ability to deal with data is nowhere near the bear that we used to deal with in the past.  The better news is that if the database design or the data being tracked needs to be significantly changed then you can clean data as you move it into your new environment.  It is OK that you weren't perfect on your first attempt and people always appreciate getting the opportunity to attack CRM after they have knowledge of what it actually is and the value to an organization.