Have you ever had a problem finding information on a website or app?
If so, you were experiencing a poor information architecture (IA). Conversely, a great experience with a site or application is only possible with solid IA under the surface.
While information architecture doesn't get the headlines that personalization or chat bots receive, when built on a unified information strategy, IA can improve the overall customer experience (CX) in several ways.
Defining Information Architecture
Information Architecture is the practice of structuring, organizing and labeling information to bring order and understanding to information products and experiences. It is about making sense of the messes. IA happens at the intersection of content, context and people.
Doing IA well means digging in to learn and apply the most appropriate structures and labels for information, developing important tools like taxonomies, ontologies and data models — it is so much more than wireframes and site maps.
Several books dive into the discipline of Information Architecture in much greater detail, including the famous “polar bear book” that started it all: "Information Architecture for the World Wide Web" by Peter Morville and Louis Rosenfeld (now in its fourth edition). The Information Architecture Institute is another great source to learn more.
4 Areas Where IA Improves the Customer Experience
1. Better User Experiences
No matter what anyone tells you, IA — not pretty design — is the foundation of good user experience (UX). Users should not have to think about how to use your site or app or content. This requires clear labels and structures for navigation, actions and outcomes.
IA also improves findability. So much content is created but never found, which is a waste of time and resources. Better IA helps connect related content and surface it to the right customers at the right times. This benefit applies to both internal or site-specific search applications as well as search engine optimization.
And here’s a fact: You can have great IA and great designs. The two are better together!
2. Improved Systems Integration
Organizations often have siloed applications and data stores, filling with information from customers via many channels. Some have ventured to connect their systems for improved consistency and reliability with mixed results.
Ad hoc connections through point-to-point integrations are fragile and expensive. Doing full service-oriented architecture (SOA) can be disruptive and costly, and sometimes feels as if it will never get done.
There is a better way: enterprise information architecture lays the foundation for smart systems integrations.
Part of an enterprise IA is developing and maturing a common data model, which includes master data and metadata standards. This can take many forms, including corporate and product taxonomies, controlled vocabularies and even full ontologies. From abstract to specific, the goal is to describe the processes, products, customers and other “entities” in the organization and their relationships in a way that is consistent and reusable across systems.
With this IA, it is possible to build out SOA in phases to connect systems through better APIs (application programming interfaces) and microservices. Connectors can be developed for existing systems, and new systems that are acquired or built should leverage the IA foundation.
Imagine a more consistent view of your customers across systems. And imagine a more consistent experience for those customers through invisible integrations. These are benefits enabled by IA.
3. Simpler Multi-Channel, Multi-Device Publishing
Content is created or captured from a variety of sources. Then it must be cleansed, edited and packaged for multiple channels and devices. Content publishers understand that single-source publishing is more efficient and cost-effective than reworking and republishing content across the different channels and devices.
IA helps with this challenge.
Consistent, semantic labels applied to content makes automated packaging and delivery much simpler and more reliable. This requires having the right taxonomy in place for categorizing content, and it is best to apply labels as early in the lifecycle as possible.
A variety of tools exist to help apply metadata consistently at scale, which enables much better multi-channel, multi-device publishing. And that leads us to artificial intelligence.
4. Smarter Artificial Intelligence
Artificial intelligence (AI) depends on IA. Machine learning and neural networks can already do some pretty amazing things, but these technologies need good data structures.
Just as humans learn by going from abstract frameworks to more specific and concrete ideas, AI applications need frameworks from which to operate and learn and grow.
For example, advanced content analytics tools can examine enormous stores of content and infer meanings and customer sentiment. An IA ontology can improve the process and narrow the scope of what the AI tools should look for.
AI can also be used to apply metadata to large volumes of content, and it helps greatly if that metadata is prescribed by IA. Even AI-powered chat bots benefit from having a frame of reference from which to work. Letting these applications learn without structured guidance leads to some pretty bad results.
Can’t AI applications examine big data stores and determine information architecture for you? Some AI tools can extract terms and suggest taxonomies — but these still need to be verified by domain experts and information architects.
With a solid IA in place, the AIs can go do what they do best — process enormous amounts of information and deliver relevant content to customers automagically.