For better or worse, I have been in the ECM industry for over a decade and a half. During this time, I have consistently seen opportunities for organizations to use ECM to bring about transformative value. But more often than not, ECM implementations fall short of the initial promise.
These so called failures are of three specific types:
So the obvious question is: What are the key causes of such high failure rates? The other side of that question, of course, is: What can be done by organizations to avoid them? A more fundamental question is whether ECM is even worth pursuing, given the odds of success – which, for purposes of this discussion, I am going to assume is not worth debating, since many organizations have in fact been successful at it; it’s just that certain proactive measures that are critical to ensuring success need to be addressed.
But before we do any of it, I think we need to understand the key reasons ECM implementations fail.
1. Disregard for Adoption.
Most organizations are so focused on trying to get something in production, that they tend to disregard the simplicity and fluidity of user experience that is required to garner the appropriate adoption rates. Adoption is a two-dimensional problem. The first dimension is making sure that a critical mass of users is enabled to access and use the ECM system. The second dimension is that a critical mass of content be available through the system. If either of these conditions doesn’t hold true, overall system adoption suffers. However, for both the number of users and amount of content, simplicity of user experience is the most critical dimension. If a user has to perform additional work to use the ECM system, that he/she didn't have to do in their old world, more than likely than not, they will bypass the ECM system. Not placing sufficient emphasis on both of these dimensions of user adoption is one of the biggest reasons ECM implementations fail.
Recommendation: Focus on scaling the number of users and the amount of content with an acute emphasis on simplicity of user experience. Integrate ECM's back end with social computing front ends.
2. Picking the Wrong Business Scenarios.
Most organizations don't pick the right business scenarios to start with to demonstrate the true potential of ECM. Those organizations that choose the first few scenarios for roll-out based on a combination of strong ROI potential and strong strategic enablement of long-term business directives can drive high (and positive) visibility for their ECM initiatives. It is the demonstration of the transformative potential of ECM to a business that tends to get the right level of enthusiasm from the other business areas.
For example, if your company is looking to move non-core business processes to a cheaper geography via business process outsourcing (BPO) activity, digitization of content is a necessary pre-requisite and enabling condition. Enabling digitization for BPO will have a higher likelihood of obtaining executive attention than digitization to save file-room real estate.
Recommendation: Pick business scenarios to start with that: (i) have high visibility at senior levels of management; (ii) offer strong potential for cost reduction or for increasing the top line; (iii) need ECM as a strategic driver to their business rather than a toolset from IT.
3. Over-customization.
Doculabs consulted for more than 500 companies across a multitude of highly regulated and complex industries, including life sciences, financial services, consumer packaged goods, utilities, state and local government, etc. Very seldom have these companies recognized the similarity of their operations to those of their peers. Instead, what we hear more often than not is how unique each company is, despite the fact that very few are that different from each other, especially regarding the reasons they could potentially fail with ECM.
One result of this conviction of uniqueness is that as businesses implement ECM, they tend to over-customize their implementations, believing they are different from the rest of the world in the ways they use information. The net result: long, arduous, risky, and expensive roll-outs, with even more expensive upgrade cycles. If most organizations focused not on meeting 100 percent of their needs with ECM, instead aiming to get 70 percent of their most important needs fulfilled by ECM, the cumulative benefit over a period of time would be far greater than trying to boil the ocean all at once. This seems like common sense, but is very rarely practiced.
Recommendation: Focus on less custom development; place greater emphasis on configuration.
4. One-off Provisioning of Business Requests.
The period from the time a business request comes in to IT to the time users have a functional system is, on average, 12 to 15 months of elapsed time. This extremely slow turnaround time is characteristic of many companies and is primarily the result of the lack of a repeatable implementation process.
Recommendation: Implement ECM as a shared service so that repeatable, configuration-based roll-outs become the norm.
5. Lack of Focus on Proving Business Justification.
A big reason why ECM implementations fail is a lack of focus around quantifying the ROI for ECM. Defining concrete and credible benefit streams and tracking against the model to ensure that the implementations produce those results is the only way to obtain sustained executive support and funding. Most business justifications tend to be either too optimistic to be taken seriously, or too generic to be applicable to an organization.
Recommendation: Develop specific business justifications with organization-specific data and a defensible heuristic model, showing peer group benchmark data that can be demonstrated to an executive in 10 minutes or less, but with rigor in the back end to be able to support the conclusions. (See Doculabs' white paper for more details.)
6. Ignoring the Impact of User Experience.
Eventually a system either succeeds or fails based on how it appeals to the user emotionally and whether the user is able to imagine the possibilities of a better world as a result of sustained use of a system. Since ECM has such a huge platform component to it, many organizations fail to get clarity in how "a day in the life of" a user will be impacted a result of ECM.
Recommendation: Build out storyboards or prototypes depicting how a user's life would be materially impacted for the better as a result of using ECM capabilities.
7. Lack of an Overall Vision for ECM.
Most organizations don't tend to have a sufficiently comprehensive vision of ECM to be able to determine whether their efforts are a success or a failure. A strategy for information management within an organization is largely absent in most places.
Recommendation: Either develop an ECM strategy, or dust off and refine your existing ECM strategy, especially in light of the rapidly changing landscape of social computing, information governance, and cloud-based computing models.
8. Underestimation of Cultural and Change Management Implications.
Last but not least, most failures aren’t the result of technology issues. ECM is a mature technology; most ECM suppliers are in their tenth and eleventh major versions of their products. The primary reason for failure is that organizations grossly underestimate the cultural and change management implications related to automating certain business tasks. People have become accustomed to doing things a certain way, and this sort of change can sometimes have a jarring effect. If not properly positioned, the ECM initiative starts off on the wrong foot and never gets back on track.
Recommendation: Whatever you're budgeting for change management during a roll-out of an ECM project, simply double it. No matter how much you think you've overestimated the effort, trust me when I say you've probably underestimated it by a long shot!