Guest post by Carl Frappaolo

I have chronicled and grown with the ECM market for many decades, in awe of the way it has progressed from simple, stand-alone file management and imaging systems, to today's complex, multi-technology ECM platforms. But I speak of the leading edge of ECM, solutions targeted at large enterprise-wide implementations. I recall a time several years ago (two financial crises ago), when these ECM players spoke of targeting the small-medium (SMB) market as a way to grow revenue and market share. Despite the move by some to offer SaaS-based versions of their products, for the most part the ECM market is still very much divided.

Looking at the market from the SMB perspective, it is not a mirror image. In fact, recently I thought to myself that ECMers that are in SMB businesses may often wonder: "What are these industry analysts talking about?" Despite the fact that the challenges and requirements of an SMB organization are pretty much the same--albeit it on a different scale--solution alternatives and product features are not the same.

The aforementioned move by ECM platform providers to SaaS models still did not conform to SMB pricing and complexity expectations. But, a whole new family of products did emerge in this same time frame, developed "in the cloud", and/or sprung from the advent of web and enterprise 2.0 technologies and approaches. These products offered simpler, more agile and less costly ECM solutions.

Despite some very attractive functionality in the form of collaboration and portal design, however, these products have gaps in functionality, functionality that the enterprise ECMer has come to take for granted.

Let's start with the first and most fundamental: Search. I have often opined that the ECM world will forever be in search of the perfect search tool--the holy grail of ECM. As discussed in a previous article, search should be fine-tuned to specific situations.

Search is fundamentally critical to a solution, and although not a substitute for tagging and taxonomy, it is a powerful complement to tagging and taxonomy that eases the burden placed on the user. The search market has morphed several times over in the last few decades, provisioning highly adaptive and sophisticated approaches to natural language processing and collective intelligence. But amongst the SMB-ECM vendors, the predominate approach to search is integration with Lucene. This is not a knock on Lucene.

Lucene is fast, reliable and accurate, but it lacks some of the add-on features and capabilities that emerged over the last few years that position search engines far beyond "simple" search and retrieve. Enterprise search veterans such as myself have come to expect customizable and heuristic relevancy ranking, social voting and integrated intelligent language handling. But Lucene, at least in the way it manifests in SMB ECM products, lacks these features and functionality.

The search experience is a throwback to the 1980s when users were expected to do a bit more digging within a search-returned result set, and to think far more carefully about the exact words and phrases they were after. While integration with a state-of-the-art search engine may be possible, it is not practical, as the SMB ECM systems are not architected for simple plug-and-play replacement of the search and, and the cost of doing so makes this alternative impractical.

This situation could be somewhat minimized with the integration of a well architected taxonomy, but, unfortunately, among most SMB-ECM products, underlying databases do not support structured approaches to tagging--facets and matrix taxonomies. Here again, the experience is reminiscent of the 1980s and earlier, when keywords were popularly promoted. Yes, you can capture tags and associate them with a body of content,  but there is no straightforward way to manage the input and quality because they are all lumped into one general field known as keywords. As a result, there is also no way to treat each tag separately--to perform parametric searches and/or intersect search results by tag values.

OK, not all SMB-ECM solution providers take this approach to search and findability. Some offer very sophisticated search capabilities and interfaces. But, unfortunately these products typically ignore fundamental document management (i.e., check/in out, revision control etc.). Among these findability-focused SMB-ECM providers, the response to the need for document management is typically: "Just use Sharepoint for that." Just? I do not even want to go there.

Lastly, there is the issue of process automation and control, aka workflow. Among some SMB-ECM products, workflow is provided, but again circa 1980. Routing architectures are limited, conditional logic often unsupported, interfaces elementary and audit trails basic.

Despite the size of an organization, the complexity of content lifecycle management, social computing and findability are not radically different between organizations. Compliance is still compliance, content is still content, process is still process and the complexity of findability is still complex.

All ECM users, no matter the size of their organization (except for very small single-site companies) benefit from an integrated approach to structured tagging and taxonomy construction, document management, intelligent natural language search, full-lifecycle content management, social computing and strong underlying database. Yet SMBs are forced to choose between products strong in either findability, content management, or collaboration--and to accept a nostalgic approach to the rest.    

Carl Frappaolo is a veteran thought leader in the ECM, KM and innovation management space. He has published four books, countless articles and whitepapers, and is a blogger at TakingAIIM. He is also a frequent keynote speaker on these topics. He is currently the Director of Knowledge Management at FSG Social Impact Consultants.   

Related Articles:
Is it time to start thinking ECM-as-a-Service? 
Today's case management is (mostly) yesterday's transactional content management  
A closer look at where BPM and ECM collide  
Give the right information to get the right ECM solution