Free-Form Taxonomy

When designing Content Management System for an organization, the architects inevitably spend time designing the Taxonomy for the content. It is enormously important from the business stand point to organize the content to be able to logically access it the way we would find paper based files in the huge locked away cabinets (this very line exposes the disconnect).

Now, I am here to question the ROI on this.

Typically, in the first Go-Live, when the users learn about Content Management Systems leaving behind their ever-reliable paper based processes, taxonomy or a folder navigation to find content would seem like a boon to users. But with time, the amount digital content starts growing exponentially [especially with legacy conversions on top of the new additions] and the folder based navigation inevitably takes a back-seat – think about it – with the advent of Google-like technologies, search has become very efficient and accurate. The possibility of someone having to navigate the folders to get to the right document is just time consuming. And with the growing amount of content and with high performance Systems becoming very affordable, search just becomes an obvious solution. How often you search your computer for the files you were working on just a week ago? And with most Systems equipped with “full-text search” capability [where in the search examines all of the words in every stored document as it tries to match search words supplied by the user], search becomes all the more important to get to the right version of the document which is just not tagged with the keyword but also carrying the keyword in its content.

Switching gears – in the paper-world, taxonomy makes complete sense.

Lets take a health-care scenario: A patient will have appointment related documents, visit related documents, lab reports, billing information etc. It makes complete sense for the health-care to gather all the patient related information within the relevant time frame [or may be all his history] when the patient is visiting. In the paper-world, you need to organize this data into charts (or folders) and have it filed in lockers based on medical record numbers OR date/time of visit OR Last name OR whatever is relevant for the health-care to get to it in a reasonable time frame. That’s the way we’ve always done it.

Now, in the digital world, searching the System for a given patient information should bring just all of it anyway. And even better, the out-of-box functionality of most Content Management Systems have the capability to present such collection in the order/format user needs it. Now, tell me, why are you navigating through all the folders?

You could always argue that –
Argument) Its more intuitive.. you’ve been organizing digital data into folders since the advent of OS..
Answer) Sure, its easy to get to data when the data is “limited” and visited regularly. But imagine when the amount of content is in millions – not that you couldn’t get to it using folder navigation, but why would you, when you could search for it in a fraction of a second?
Argument) It would be required to use folders to define a certain business functionality
Answer) Why not just use tagging or attributes? After all, folders are just references, right?
Argument) It gives yet another way to get to data besides search, especially when you suspect that search is not accurate.
Answer) And yes, that’s what its real worth is – not the primary form to get to the content, but a back up. And this needs to be kept in mind when designing it.

Of course, taxonomy definitions could be matter of compliance and I am not preaching against it. Compliance is good and there’s a reason why its in place. But the point I am driving at is to understand the real-worth of the taxonomy. It is the “back-up” solution to get to the content – not primary. And this needs to be firmly kept in mind when allocating time/resources to designing it.