A blog series focused on crafting meaningful structure in your Digital Workplace
Part 3 of our series focuses on tactics to organize the user insights you've collected.
Up to now, you’ve been gathering insights that will support an effective information architecture (IA), but now it’s time to dig in and start considering tactics for building your IA, including organizing all that information into a formal structure.
With the content inventory complete, you have a good idea of the information needed in your digital workplace. But how should it be organized and connected for the easiest, most intuitive access? To get further input from users, you can use some tried-and-true tools that just happen to be analogue:
1. Card sorting
Card sorting helps you figure out where users would want and expect to find something in a digital workplace. It’s a low-cost, simple technique that involves labelling index cards with different pieces of content or functionality, and then asking users to sort them into groups that make sense to them.
There are several types of card sorting methodologies. The basic method starts out with cards in random order and users sort them in the way they think they should be grouped. In reverse card sorting, the cards are pre-sorted into groups, and users are then given the task of rearranging them as they see fit.
In the end, card sorting helps you gather ideas from users on how content should be grouped and labelled.
Whiteboards are another analogue tool commonly used during the initial planning stages of IA. Whiteboarding makes it easy to visualize content, quickly record ideas, and make connections with a team.
There are also a number of digital IA tools – such as virtual whiteboards – that are growing in popularity as more organizations have teams that are distributed or working remotely.
Choosing an organizational structure
Once you’ve evaluated the output from card sorting and/or whiteboarding sessions showing trends in grouping and labelling, you must decide what kind of structure to draft.
There are two main approaches for defining an IA:
1. Top-down IA structures
This approach takes all the knowledge you’ve gathered about business strategy and user needs and creates a basic, high-level structure. From there, you work your way down to create the detailed relationships between content as the architecture deepens.
Taxonomies (hierarchical structure of content) are often used as tools for organizing content. In a complex system like a digital workplace, taxonomy also often refers to metadata and controlled vocabularies.
Metadata describes a piece of content and provides a meaningful set of attributes that can be used further classification.
Controlled vocabulary is a restricted list of terms used for indexing or categorizing.
2. Bottom-up IA structures
This method focuses on the detailed relationships between content first, examining how the digital workplace can facilitate specific user requirements. Then you consider how the high-level structure will support these requirements.
One example of a bottom-up organizational structure is the Database Model, which creates a more dynamic experience for users. It allows for filtering and advanced search capabilities, and provides links to related information that has been properly tagged.
The best IA looks at a site from both angles, and uses different organizational structures in a complementary way. Projects that ignore top-down approaches can result in well-organized, findable content that doesn’t meet the needs of users or the business. And a project that neglects the bottom-up approach can result in a site that makes it easy to find information but doesn’t promote discoverability of related content.
Translating your strategy into IA blueprints
There are several methods for capturing and defining IA. All of these techniques are aimed at helping your stakeholders visualize the structure and provide feedback at an early stage, before any designers or developers are involved.
1. Site mapping
Sitemaps are quick, inexpensive ways to visually represent how different pages and content will be grouped, what order they’ll appear in, and how they’ll relate to one another. They are essentially high-level diagrams of the overall content structure, but not necessarily of the navigation structure. Sitemaps can’t be finalized until page layouts are defined, which leads us to the next technique.
Wireframes are annotated visual illustrations of page layouts. They communicate page-level navigation, content types, and functional elements. The annotations are for designers, so they can mock up a visual interface, and for developers, who need to understand the page features and how they’re supposed to work.
Paper-based wireframes can be a good starting point, but digital versions are often better for collaboration. Once a wireframe starts to have interactive elements such as clickable navigation, it crosses the line into a prototype, which allows you to test a site before coding and development begin.
All the research and conceptual design we’ve talked about so far has been focused on a top-down approach to defining a structure that will accommodate your business mission, users, and content. Now, moving into production, you step into the bottom-up phase. Content mapping is where top-down meets bottom-up.
In content mapping, you separate all your content from its existing containers, breaking it down into logical ‘chunks’ that merit or require individual consideration. As you’ll have learned in the content inventory, your content comes from a variety of sources and is in a multitude of formats. All this content must now be mapped within the IA. Separating content from containers is essential, because it allows it to be used across multiple pages.
To define your content ‘chunks', ask these questions:
- Can this document be segmented into multiple chunks that users might want to access separately?
- What is the smallest section of content that needs to be individually indexed?
- Will this content need to be repurposed across multiple documents or as part of multiple processes?
Once you have your chunks, it’s relatively simple to map them onto destination pages.