Some time ago I was at a Microsoft Convergence conference and was immersed in all things CRM. During one discussion the topic of the costs of CRM implementations came up. I distinctly remember one of the comments.
“The trouble with CRM projects is that the first 80% of the project takes the first 80% of the budget and the remaining 20% of the project takes the other 80%.”
I recall that we all had a pretty good laugh over this.
Like most humor, it’s funny because it’s true. The typical CRM project costs more than originally expected.
I think there are a few reasons for this. The first is most of us are unreasonably optimistic and tend to underestimate the complexity of the average project. This just doesn’t apply to CRM projects – it’s virtually every technology project. (See Management by Wishful Thinking for background on why this occurs.) Invest some time and though in the planning process in order to combat this.
A second reason is that companies try to go it alone and don’t get expert help. It is very difficult to get it right the first time; contracting some expert help can minimize the risk greatly.
A third (and biggest) reason is that the scope of CRM projects become broader than originally planned. There are several factors that influence this.
Business Processes: Your company probably has its business processes documented and there’s a good chance that they reside in some dusty binders resting on an obscure shelf somewhere. The challenge is that they are likely out of date. Processes change fairly regularly and unless your company is extremely diligent the documentation doesn’t necessarily get updated. And these are just the formal changes. Users develop their own shortcuts and workarounds and these never get documented. The processes in those binders are at best an approximation of what actually happens. So you are going to have to spend some extra time understanding the current state of your processes.
Organization: It’s not just business processes. To get the maximum benefit from CRM you have to make sure the culture and philosophy of the organization are themselves customer centric. Companies that are implementing CRM for the first time typically aren’t focused on the customer – they are focused on whatever product or service they offer the customer. The journey to becoming customer-centric is longer and harder than expected.
Another organizational consideration is who owns the customer? I have seen CRM create turf wars in organizations over ownership of the customer. Sales? Marketing? Customer Service? Does the regional office or national office own the customer? CRM requires a customer engagement strategy where this type of thing is defined in detail, or else your customers could be bombarded by disconnected messaging from various functions in your organization. Developing this strategy takes time and money.
People: Winning the hearts and mind of the users can be difficult. It’s great that CRM is good for the company, but what’s in it for them? There will be resistance because the implementation of CRM is creating a new area of subject matter expertise and likely diminishing the importance of some existing subject areas. That’s a difficult situation for the present subject matter experts. Change is always hard – it can be helped along by providing compelling explanations of the benefits that the users will get from CRM. Make sure that end-user training is complete and answers the ‘why’ questions as well as the ‘how’. Also remember that CRM can give you a lot of information on end-user productivity and that attention is not always welcome. The bottom line is that training and education will be a larger effort than you originally anticipated.
Integration: Most CRM systems are integrated with existing systems and data. Integration with existing systems is harder than you think it will be. There’s a number of reasons for this. It’s likely that, like your business processes, your current systems and data are inadequately documented. The effort to better understand your current state will again be larger than expected.
Once you understand what you have there is the issue of data quality. Data quality has numerous properties that need to be considered (accuracy, validity, timeliness, consistency and completeness). The older your data the greater the likelihood that a data quality effort will need to occur before this data can be integrated into the CRM system. This is not a trivial effort.
One more thought about integration – there are actually two integration efforts that need to be considered. The first is the initial data load, the second is the periodic (daily, weekly or whatever) update/refresh. The efforts are similar but different; plan to manage them as two separate and distinct efforts.
Security: CRM collects (and generates) a *lot* of data. It needs to be managed and safeguarded. Regardless of whether it’s on-premise or in the cloud, security, confidentiality and privacy of data is a big deal and by implementing CRM you have just made it bigger, particularly if any of the data is being exposed to the internet through a portal. Protecting the CRM data from unauthorized access from both internal and external sources needs to be considered part of the project.
Success: Success can be a problem. Once the business community starts to better see the benefits of CRM there will be pressure to broaden existing functionality or add additional capabilities. There will be demands to broaden scope and start adding these additional capabilities immediately, particularly if you are implementing CRM in a series of small releases. Indiscriminately adding scope to the project is an easy way to turn a 12 month project into one that takes 18 months. Ensure that you have a robust change control process defined or you may find yourself with a project that never ends.
CRM is worthwhile investment but it’s good to start with your eyes wide open. There are aspects to CRM projects that aren’t readily apparent at first glance; missing these can lead to increased costs, extended durations, unfulfilled expectations and general disappointment.
And who wants that?