A speculative look at where Information Architecture might be going, given machine learning and voice interfaces.
“person opening the book” by Marten Bjork on Unsplash
What is Information Architecture?
Information Architecture (IA) originated from library and information sciences — it’s the study of how information is created, managed and organized.
What makes good IA?
- results in a structure that makes sense and is easy to navigate
- focuses on organizing and labeling websites so that users can best find what they’re looking for
- depends on the interplay between the meaning of the product’s elements (ontology), the arrangement of its’ parts (taxonomy), and the interaction among its’ parts (choreography)
On a practical level, this means making decisions around organization, labeling, search, and navigation.
In the context of IA, ontology focuses on labeling and organizing, taxonomy focuses on organization, and choreography focuses on navigation and search.
There are three types of navigation that choreography focuses on:
- global navigation- visible on every single page (e.g. the shopping cart when you are on an e-commerce website, or a share link option when you are on a social media platform)
- local navigation- immediate area/category/task
- contextual navigation- shows users similar entries or associated content (e.g. the “customers also bought” section on Amazon product pages)
IA Today
90 percent of the data in the world has been created in the last two years alone.
What does this mean for designers? The role of an IA has changed over time. Now we use the term (UX) user experience designer, or product designer, and these professionals have a broad set of information architecture skills.
Understanding IA is important for anyone designing products that people interact with in order to achieve a goal or task or digest some kind of information. It is up to designers to help the user navigate through an information space in search of something they need, as quickly and easily as possible.
So how do designers determine whether IA is effective or not?
IA Heuristics
IA Heuristics is the evaluative criteria that designers use to determine whether IA is effective or not. There are ten ways designers can measure the effectiveness of IA:
- Is it findable? Can users easily locate what they’re trying to find? Is there more than one way to access things? How is findability different across devices and platforms?
- Is it accessible? Can it be used via all expected devices and channels? Is your product resilient and consistent across channels? Does it meet standards of accessibility for its’ target audience? Does it consider users who are visually or hearing impaired?
- Is it clear? Is it easy to understand? Is the target demographics grade and reading level considered? Is the path to task completion obvious and free of distraction?
- Is it communicative? Are messaging and copy effective for users to complete the tasks at hand? Do the navigation labels and messaging help a user orient herself within the product? Are labels and messaging consistent across the product and its’ channels?
- Is it usable? Are users capable of producing the intended result? Are users able to complete the tasks they set out to without frustration?
- Is it credible? Is the content up to date, and updated in a timely manner? Is it easy to contact a real person? Is it easy to verify the products security when making payments?
- Is it controllable? Are tasks and information a user would want to accomplish available? How well are errors anticipated and eliminated? When errors do occur, how easily can a user recover? Are features offered to allow the user to tailor information or functionality to their context? Are exits and other necessary controls clearly marked?
- Is it valuable? Is it desirable to the target user? Can a user easily describe the value? Does it improve customer satisfaction? Does it create enough value that users want to pay for it?
- Is it learnable? Is it able to be easily understood and used? Can it be grasped quickly? How does it reduce the complexity of more complicated processes? Does it behave consistently enough to be predictable?
- Is it delightful? How is it superior to competitors? How does it differentiate on the same features? How are user expectations exceeded? What can you turn from ordinary to extraordinary?
So now that we know how designers measure good IA, how does this fit into machine learning and voice interfaces?
Artificial Intelligence (AI)
AI is the theory and development of computer systems able to perform tasks normally requiring human intelligence.
“2nd. gen black Amazon Echo speaker on white panel” by Jan Kolar on Unsplash
The challenge is making this useful for users and ensuring that the suggestions are relevant, in the correct tone and understand the context of the message.
There are two ways for voice input — voice first devices (e.g Amazon’s Alexa) and screen first devices which has integrated software (e.g Siri).
Voice interfaces provide some great opportunities for designers, IA’s and users. However, there are some challenges:
- How can I use it? — Since there is no visual platform telling you what to do, it’s not always clear what actions you can perform.
- Computer rejects you — voice interfaces have a tough time deciphering content since one word can have multiple meanings. This results in a frustrating experience for the user.
How does this affect information architecture and designers?
- We as designers need to make sure the content is findable and usable by everyone, regardless of their ability and regardless of the device their browsing on.
- How we design voice interactions and conversations to fulfill the needs of users to complete tasks, is now much more important to help the user find information and solve problems on a more intelligent level.
- Designing a product with no interface and still making it as usable and inclusive so that it’s clear what it does and how to use it, will be a new challenge for designers to overcome, but will also help invoke more creative solutions.
/w=3840,quality=90,fit=scale-down)
