A couple of years ago I was involved in a fairly large CRM deployment for a company I worked for. My step-brother was working for Siebel at the time and had some very wise words of wisdom for me…
The failure rate for CRM deployments increases proportional to the amount of customization performed.
I have actually found this to be true of any large system deployment, be it CRM, ERP, Financials, HR, etc… What is interesting about this observation is that most large application suite comes 80-90% finished. The rest is left for the vendor/integrator/customer to customize for the specific installation. So does this mean that all installations have an automatic 10-20% failure rate? Not necessarily, you see every vendor and every software package comes with a pre-determined set of assumptions on how the package is going to be used. These assumptions are fundamental to the design of the system and based on how much deviation there is from these assumptions will dictate the failure rate.
So with this in mind any company will be confronted with the first decision to make. Build or Buy? This question and decision is really comprised of a series of sub questions and decisions, i.e.:
- Ask yourself if given the specifics of your company and your needs for the application is there a vendor solution that matches exactly what you are looking for and how your business operates?
- If not, how different is the process workflows?
- How much would it cost to customize the application to mimic your organization’s process workflow?
- Would that customization cost more or less than building a system from scratch?
- How long would that customization/creation take, is it worth more given the timeline to go one direction vs. another?
In my opinion there is another question that should be asked, that often is not…
What would be involved/what would the cost be to align the process workflow of the organization to the workflow inherent in the leading application suite? Is this cost less than the customization/creation of a system to match your organization?
Any vendor worth evaluation for a project like this has done a significant amount of market research, and through this research and their previous deployments has identified the best practices for any particular workflow. So their application is going to be tailored and focused to these best practices, meaning that if you really have that much deviation you are pretty far away from what the best practice is… is that where you want your organization operating? Sooner or later you are going to have to adjust your business, why not sooner?
More on this thought will follow…