How to read the Information Architecture
Whatever a user approaches, instantly she has a vague idea of what she can find there or expect. It drives from System 1, instinctive, guts and experience-based presumption. User creates a mental model. A model that interprets symptoms into explanation — for user’s cognition sake — of how something operates, behaves, what consequences and results are.
Imagine, each time you approach a computer mouse, you have to learn its operational rules. Indeed, — but for its affordance and your experience — you know immediately how to manipulate it, and what results it bears. So, you do have its mental model. You can tell, it translates its position and movements by some kind of sensor and passes it over to the computer.
Digital products aren’t palpable, but besides touch, all other senses play a role in interaction with them.
Information Architecture (IA) conveys affordance for what kind of knowledge is available or what goals are accomplishable with the product (may it be website, app or software).
Building Information Architecture
When building a successful Information Architecture;
- The first step is to investigate what we have. What the knowledge is, how it is presented, what it is composed of, who owns it, what are artefacts, how they relate to each other. It is about structure and schema of implementation model.
- The second is to understand user’s mental model. That means how she thinks about the available knowledge, what she expects to achieve and how she is going to do it.
- The third step then is to build a model that will translate the knowledge into digestible chunks of information to a user. To build representation model that drives from implementation model, but meanwhile it fits user mental model (her expectations).
Product’s models
Information Architecture (IA) is causally connected with a representation model. Implementation model is raw data, business goals, organizational goals. It is data prepared by developers and business people. It gives some information, but it is not particularly directly the same as knowledge, on user's part, what is a mental model. The mental model represents the knowledge or presumptions users have on data and goals and how to get them. A representation model is the information about data and presented in meaningful way to the users in the way it teaches her what goals can be attained and what knowledge is available. It instructs the user how she should think about the product.
UX Designer is responsible for the IA of representation model. It is composed of 4 systems.
1. Organizational System (the backbone). The high-level structure of knowledge, kind of knowledge, relation between information chunks, user goals, organizational goals.
2. Navigational System (the way of getting around the informational environment). The browsing, finding, clicking through.
3. Retrieval System (search, wizards, configurators). Guided user paths, user flows allowing for finding the knowledge, finding artefacts, achieving her goals.
4. Linguistic System (content, semantics and its vernacular language). How the user is being told about the knowledge, how she is being informed about achievable goals, how she is being encouraged or how things are explained to her.
1. Organizational System
Organizational System is the backbone on which UX designers build their IA. It determines building blocks, their size, mutual relation and interdependence. It is rarely changed, and it falls out of designer’s competence to redefine it. But proper understanding of implementation model is the first step to build the IA. The way the data is organized implies on how it may be presented to users. Furthermore, when designing from scratch —it determines how the data is organized in the back-end.
The data (in implementation model) is all what is on the side of business, organization, and developers. Business processes and goals, relational databases and sets of rules, organization structures and departments, records, scores, big data and logs, APIs and third parties. One may think of it as a “back-end”. The hidden part. Everything that is required for product functioning but is of no interest for the user. It would be a technical CAD drawing or mouse and schema of circuits.
UX designers are obliged to make an inventory of this data, item clusters and their mutual relation and interdependence. Understanding what are key actors, and what we know about them (data about data) is the clue of their successful representation. Between these items, there exist three ways of relation. Relation by hierarchy, by metadata or by human decision.
A. Hierarchy
Because of the pervasiveness of hierarchy, users can easily and quickly understand information environments. They are able to develop a mental model of the environment’s structure and their own location within the structure. The representation model, based on a well-designed hierarchy is SEO and bots proof, and works perfect for non-human interaction.
Hierarchy, being vertical, can be either deep or shallow, broad or narrow. How we navigate through it (Navigational System) depends on hierarchy. The best approach is always the golden middle. Narrow and shallow rather than broad and deep.
Hierarchy is imposed from above, it builds on organization schemes (defines the shared characteristics of content) and organization structure (defines the types of relationship between content item and groups).
Organization Schemes. We acquire understanding by experience. Imagine finding Organic coffee grains in a large, unfamiliar supermarket. Is it close to the tea, colonials, or maybe in Organics?
Organization schemes can be exact (like old time, without computer alphabetical indexes, geographical and chronological), ambiguous (decided on shared characteristics, like topic, audience, task, and metaphor) and of course mixed, hybrid schemes.
B. Metadata (database model)
It is a bottom-up hierarchy that allows for flexibility. It stores rich data (characteristics, metadata) about items but creates no fixed organizational structure. Items can be rearranged instantly, displaying any kind of structure, relation, or lack of structure, you want. It is not impossible, but very difficult, to add a new characteristic (organization scheme) to existing items, allowing for re-structuring of the organization scheme.
C. Human-based hierarchy (folksonomy, hypertexts)
It is a strikingly subjective organization, reflecting what its creator judges relevant and worth of linking. The system can be like a neuronal wire mesh of links, without any item above, where each node is connected to several others. It can also be a hierarchy created by a cohort of users, and their judgements translate not directly to the relation, but via a voting and counting system (appreciation) or free tagging (social classification, folksonomy).
2. Navigational System
Navigational System deals with how user accomplishes her goals. How she is guided towards completion through all organization’s available items. It is composed of navigation bars but also wizards, configurators, call-to-action buttons and direct paths to particular goals. The main task of the Navigational System is to guide user through red routes (user goals execution); facilitate fast understanding of the information environment (what data is available, what goals are accomplishable); allow for browsing and easy user’s self-localization (discovery and findability).
- Global Navigation Systems. Permanent and visible on every screen, that also sets the information context.
- Local Navigation Systems. Ribbon, section, side panel, toolbox, breadcrumbs.
- Contextual Navigation. Content related links, advanced menu options, task related decisions, goal related CTAs buttons.
- Supplemental Navigation Systems. Means to explain the whole content. Site maps, indexes, hyperlink, or keywords clouds.
- Guides, wizards, process steps, and configurators. Any kind of walkthrough for users, leading to one goal execution.
- Progressive disclosure. Scrolling through content, previewing sections’ subjects. Window displays walking tour.
UX designer shall understand red routes, business and organization goals, lead contextual inquiry or cognitive walkthrough to pinpoint pain-points.
3. Retrieval System
There are two roles for Retrieval Systems in IA. One is to serve as a rescue party when the user fails to discover the hierarchy pattern and find what we came to find. The second role — if the unique user goal is to find an item — is finding (it should be the first product’s thing visible).
- Recovering from getting lost. These are all systems that retain the user from leaving the product without finding his search. These are search field, but also chatbots, ask the expert and blind paths, and shall be visible when needed only.
- Finding. All aids helping user find the exact catch, refining search query, correcting mistakes, thesauri, boolean operators, searching within results, displaying additional information about results, suggesting entries, default results.
Most powerful Retrieval Systems occur when organizational hierarchy is database-driven. For it having rich data about data, it grants user with complex, multistep finding and grouping results.
If designing, UX designer shall inquiry users, how they search for the information, and what information they need, when they call it success finding; also, why they are not finding this what they are seeking and how — in other ways — the item sought for, could be presented to them.
4. Linguistic System
The first thing that catches user eye is text. F-reading pattern (or more precisely scanning pattern) assures the user sees navigational categories and headline with CTA on the first fold, reading them as tag lines. What wordings these labels bare, and how they address the user, determine the understanding of where the user is and what stance the product presents toward users. It is the metaphor of building’s facade. Once knowing what we can expect by the mere face, we aren’t surprised and adapt our information seeking patter to the digital place we came up to.
- Labels;
- Content structure (H1-H6, and paragraph);
- Call-to-action phrases and Claims;
- Emphasis
A UX Designer should understand the jargon or domain nomenclature by undertaking Card Sorting and Reversed Card Sorting; to understand how the users think the product functions, designer should do a Contextual Inquiry, the best in the manner of “Day in life”. It is also important to consequently follow UX Writing style guide, to address user persona in the same manner, and same level of formality.
Summary
Information Architecture is a system of backbone, sinews, and ties. It represents the product’s purpose, and way of usage. It also fulfils additional tasks of hinting the user about information environment pattern, presenting all available inventory, introducing domain, and allowing for associative learning. IA is most obvious in representation model, the designed representation of a product functioning. IA draws its building pieces from implementation model, but how user understands these pieces is her mental model. Designer’s task is to save on user’s cognitive dissonance between mental and representation model.
Each product can be divided down to its Information Architecture. Whatever the face is built onto IA, it is a firm and unchangeable system that warrants users’ success. Each product’s IA can be described by characteristics of 4 systems that are unique for domain, genre, or even product. Understanding these systems deepens designer’s knowledge on why a product earns or botches.
