Classification in the Age of Search

Last summer, when my daughter and I were visiting Pittsburgh, my GPS took us to the middle of a bridge over the Ohio River and announced “arriving at destination, on right”. The problem was the address we had been given for the Italian restaurant included a colloquial name for a road, and the GPS recognized it as the name of the bridge. In addition to realizing that our pizza would be cold, I knew I had the intro for a blog on classification. OK, maybe not classification per se, but the importance of classification. Many of the people I talk to suggest that modern day search options reduce the need for classification. I love search, just like I love my GPS, but if you don’t know where you are going, you might find the technology leading you astray.
I am not going to rant about search today; I’ll save that for when the AIIM ECM Roadmap brings us around to search as a topic. Classification, on the other hand is the topic today and I can’t emphasize its importance enough. Classification puts documents where they belong, where we expect to find them, and ironically, where we don’t expect to find them. In fact, that last quality is why we need classification in addition to search. We have to accept that we don’t always know what we are looking for, and that others don’t know our subject areas as well as we do. When I can’t find a document I created, I can search for it and find it pretty quickly; that’s because I know what terms will zero in on that document fast. When I need a document someone else created, I rely more on classification, because I don’t always know what to search for. Classification is also important when third-parties (e.g. auditors, lawers, etc.) ask for information. It is a very nice feeling to be able to say “the stuff you’re looking for will be in this library.”
SharePoint gives us many ways to support classification. Site collections, sites, sub-sites, libraries, folders (yes, I said folders) and content types all aid us in establishing a meta-identity for our documents. If you are approaching SharePoint from the technical side, like I am, you want to start building out that perfect structure right away; that would be a mistake. Classification begins with planning and should involve people who may not be involved in the creation or even the daily use of the documents being stored. When we met to begin discussing our top-level classification scheme, I was joined by five others and only one person in the room creates a significant portion of the documents we store. I was there, because I understand the technology, and the other people in the room were the ones who know how our business is organized, how we are asked to produce documents for our stakeholders and what documents define our business. Organizing those documents was easier than I would have imagined; the structure was both narrower and shallower than the one I would have built.
In reflecting on that experience, I think the reason people say search is more important than classification might be the result of bad classification, specifically “over classification”. If we work to put every document in a very small and very well defined bucket, we are requiring people to know what they are looking for. If they know that, they might as well use a search engine. Our experience tells me that broader categories and fewer levels, both leading to larger collections of relevant documents works better. Once you get people in the right general area, they are much more likely to find what they need. Think of a library. If I want a book on C# programming, I can look in the catalog (search). If I just want to see what books they have on the subject of ‘programming’, I can just walk the stacks to the right aisle.