Managing Information Like a Chemist: Creating a High-Level Information Architecture

Remember the periodic table of elements from sophomore year in high-school chemistry? It’s back.

Well, not exactly, but those months spent memorizing the properties of oxygen and hydrogen (not to mention the heavy metals) might have a renewed utility for content management professionals.

Recently, a number of financial services organizations have hired Doculabs to assist them in structuring the content used in their online customer web sites and mobile applications. Originally, the web sites were designed to facilitate a variety of different transactions. As expected, content was needed to support the transactions – instructional information, regulator disclosures, etc. As traffic grew, the marketing team recognized an opportunity, and began adding cross-sell offers – requiring still more content. Then more transaction types were offering, requiring still more content, etc. (Wash, rinse, repeat.) Now, 10 years later, the scope and scale of most financial service institutions’ web and mobile properties is significant – and for some of them, perhaps overwhelming.

Well, throughout this evolution, no one really had a plan for how all this content should be structured in the repository (or repositories) where it was stored. It was added on a project-by-project basis, often without consideration for the process or resources needed for the periodic refreshing and/or expiration of content. The result? Basically, the web content of most large financial services firms is a mess. And worse yet, with the drive toward quickly enabling these same transactions via mobile apps, the content gets ported enmasse to those new applications, where it then begins a life of its own. Now we have two messes.

What can be done to fix the mess(es)?

The best way to start is to build an information architecture. Just as many organizations normalized customer data and created a master data management plan or standardized product attributes when they adopted an ERP system, the same needs to be done for all of this unstructured content. Given the hundreds of thousands of content objects and artifacts now being consumed by web and mobile applications, imposing structure – from the very simple approaches, such as naming conventions and folder hierarchies – to the more complex approaches, such as meta-data tags and mapping hierarchical relationships – becomes critical to the long-term management of content within a collection of web sites and mobile apps.

But for many organizations, getting started at the building of this information architecture is a daunting prospect, given the size and scope of most organizations’ web properties. One of the techniques Doculabs uses is analogous to creating the periodic table of elements. Remember from your high-school chemistry class: Everything in the universe can be broken down into its fundamental elements. And there are not that many of them. So, too, can content be bucketed into essential “elements”. This is Doculabs’ approach to information architecture. And we make every effort to keep it simple.

Generally speaking, every page generated on a bank’s web site is made up of no more than 50 to 100 different types of “elements”. Unfortunately, however, there is no common lexicon for describing the “elements”, but generally they center around the categories of purpose (offers, disclosures, actions) or content types (text, video, logo, graphics). Start by digging into your wireframes or use cases (if you have them). And, just as with the periodic table, you’ll remember there was room on the bottom right for future chemists to discover new elements, implying that our existing understanding of the universe is imperfect and less than complete. Same thing with information architecture: Don’t expect perfection; not everything will fit nicely. But if you can map out 70 to 90 percent of the content that is frequently being changed or updated (or that should be changed or updated, if it were capable of being easily identified or “in one place”), it represents a huge step forward in terms of organization.

Then, in terms of implementation, as content is “touched” (updated, changed, archived), it’s enriched with structural attributes (different systems enable this in different ways). Over time, content is organized in a manner that enables better management.

The infographic here shows an example of the kind of client-specific “Table of Content Elements” we create for clients:

These high-level information architectures get business and IT professionals excited. As one client said, “Finally, I can wrap my head around the content we use in our web site without having to dig into some 20-tab spreadsheet with thousands of rows! I have the attention span of a gnat; this thing [i.e. the Table of Content Elements] is perfect.” This one little tool helps organizations get started in structuring content and creating information architectures a productive manner.

And who knows – you might even share some funny stories about high school chemistry while doing so.

Click on the image below for a full view of Doculabs Content Scientist Infographic.

Content-Scientist-02 (

Rich Medina
James Watson
I’m President and co-founder of Doculabs, serving as executive sponsor on consulting engagements for financial services clients.