Information Architecture
Machine Learning and Responsive IA: How They Can Work in Tandem to Produce A Better User Experience
The world is full of concepts and these concepts are further broken down into seemingly atomic particles. It’s the way we learn and the way we prefer to consume. Breaking something down to the bare bones only to build it back up again is often essential for human comprehension and knowledge. Our day-to-day lives are full of tiny decisions that we make in sequential order based on what we expect to come next or even two or three steps down the line.
For example, we have an end goal of cooking dinner. Based on our specific circumstances, we might understand the flow of events to be: get in the car, drive to the store, purchase ingredients, drive home, prep the ingredients, cook the ingredients, eat dinner. It’s logical, it makes sense and though it is a series of decisions, it is essentially effortless to carry out cognitively.
The funny thing is, when you put this same process in the context of the internet, things suddenly have the potential to become much more overwhelming, confusing and even irritating rather quickly. How and when we make decisions that we once considered so subconscious seems awkwardly intentional and not in our full control. Good usability and IA standards aim to close the gap between our experiences in-person and our experiences online but there is still a noticeable disconnect.
This disconnect is due to the difficulty in attempting to streamline a seemingly mundane task that just so happens to have the potential to vary significantly from user to user. Some people like to take their time grocery shopping while others like to be in and out as quickly as possible. Some like to shop around for the best price while others are willing to pay a little bit more for convenience.
The true future of digitizing tasks such as grocery shopping lies in the successful marriage of machine learning and responsive IA. Instead of companies designing for their target user, what if they could utilize machine learning to design fully adaptable IA systems/interfaces for all users. With the intel from machine learning, the future of IA promises to be one that is extensively adaptive to individual user wants and needs and it could even start speaking to subconscious needs as well.
