Verify that the information architecture of the website conforms to the user's retrieval habits

--Make it easier for users to find the information they need 1

Recently, I suddenly wanted to write a topic class article, on how to make users more convenient and quick to find the information they need on the site, of course, can not be divorced from the theme of the blog-site data analysis, so here is mainly about how to optimize the site through the site analysis to achieve the user's rapid access to information, The first introduction is the information architecture of the website.

About information architecture, web design students are experts, I am here just to swim, according to the university reading information structure of some of the remnants of the memory, plus these days cramming simple review, here to express my humble opinion.

Definition of information architecture

According to Wikipedia's definition, information Architecture (information Architecture, referred to as IA) is a composite structure that affects system organization, navigation, and classification labels in an information environment. It is a specialized subject based on information architecture methodology and using computer technology to manage and organize information. Information architecture is not applied to website design at the beginning, it originates from information science, and it should be used in the organization and information retrieval of library and other places.

Elements of user experience-user-centric web design the definition of an information architecture in this book is based on Web design: The information architecture focuses on designing organizational classifications and navigation structures that enable users to efficiently and effectively navigate the content of the site.

Concrete is no longer said, may have their own understanding, here directly to see an example--wordpress information architecture mode:

Of course, the above diagram can only show a general information structure of the Web site, the middle of the class layer may be more than one layer, there will be large classes, subclasses, child classes ... The bottom can be an article or a page or some other specific content. and the internal relationship of the site is often because of global or local navigation, the site chain and content related functions such as the existence of more complex, the diagram on the arrows will be a lot of density, but we do not need to list all the relationship between the key is to clarify the basic structure.

Types of information schemas

or refer to several classifications of the information architecture in user-centric web design from the elements of the user experience:

Hierarchical structure (hierarchical Structure)

Also known as the tree structure, is the most common web site information architecture model, the above examples of WordPress's information architecture is a typical hierarchical structure. The direction of the arrows in a tree structure is not necessarily top-down, it may be bottom-up or bidirectional, and there is a pointing arrow between the layers of content because of the presence of some associated links.

Matrix structure (matrices Structure)

The matrix structure pays more attention to the concept of "dimension", that is, from a multidimensional perspective to retrieve information, such as time, region, content classification, and so on, the typical application is Content management System (CMS) Web site or E-commerce sites, such as you can browse the Watercress film: 2010-United States-science fiction, perhaps this time "Iron Man 2 is present in front of you.

Linear structure (sequential Structure)

See the linear structure you might think of crumbs right away, and it will show the most important information architecture in the site, even if it doesn't provide you with the plane coordinates you have on the site, at least it shows you where you are on the key line, and of course, some of the key paths to the site are usually related to the linear structure, such as the user registration process or E-commerce site purchase process.

Web analytics and Information architecture

According to the different business model of the website, you can choose the mode of information architecture suitable for your website, no matter what kind of information architecture mode above, as long as the design and use of reasonable, users will be able to your site in the most convenient form, the fastest speed to find the information they need.

But when I browse some websites, sometimes really let me feel "can't find North", the result is directly close the page, if you do not want to let users have entered your site easily left, the quality of the site information architecture will directly affect the site's user experience. Therefore, we need to use some methods to test the information structure of the website to meet the needs of the user's information retrieval.

1. Try to sort out the website Information architecture diagram in the above example

This is the simplest and most intuitive way, if your website information structure is clear enough, so it is not difficult for you to draw such a picture, and when the application of the site is more complex, the content is relatively broad, then may want to sort out the overall information structure of the site will be relatively difficult, but I believe that a well-designed site as long as a little finishing, The general information architecture is still in the picture, and when you're racking your brains to make sense of your site's information architecture, your site needs to be optimized.

2. Verify the rationality of information architecture by means of web analytics

The subtitle of this article is "make it easier for users to find the information they need," so we need to analyze whether users can quickly and easily find the information they need on your site, and recommend a way to find lost usersin the site (Lost Visits) .

In a reasonable information architecture, most users will not get lost on your site, otherwise, the chaotic information architecture will lead to a large number of users lost direction, it is like entering a huge maze. So how do you find these lost users? We can analyze the behavior of this type of user first, the most obvious feature is: successive clicks on several pages, each page is only preliminary browsing (because they do not find the information they need) to go to another page or directly away. So we can use the two metrics in Web analytics:

　　Browse Pages (Depth of Visit): Total number of user browsing pages in one visit;

　　average page Stay time (avg.) time on page: the average number of times a user stays on each page in a browse, that is, the total stay (time on Site)/page of the access (Depth of Visit).

We can user segmentation methods to the number of those browsing the page, but the page average stay time shorter user browsing as a lost user, the specific value can be defined according to the characteristics of the site itself, such as I define my blog browsing page number is greater than or equal to 4, and the page average stay time is less than equal to 15 seconds of visits for lost users browsing behavior, we can use Google Analytics Advanced Group (Advanced Segment) To differentiate this type of user, on how to use Google Analytics Advanced group features, you can refer to the Blue whale's article-Google Analytics feature-Advanced group , the following figure:

Of course, you might say that this user-differentiated approach is inaccurate, that users are not necessarily lost users, or that they do find and browse specific content, but leave the page immediately because the content is not attractive or for other reasons. Therefore, the number of such visits with the advanced group can not be regarded as an absolute value of lost users, we can only think that most of the visits are lost users, and do not rule out the existence of some alternative. Therefore, a more reasonable way is to calculate the proportion of the total visits of the site visits to analyze the information structure of the site is reasonable, we can select the site in Google Analytics all visits and lost to compare the proportion and trend , as shown in the following figure:

The proportion of lost users browsing in the Web site can only be calculated by lost Visits/all Visits, but you still can't judge whether the information architecture of the website is good or bad based on this calculation, because there is a lack of a baseline ( Benchmark) or the standard of judgment. A baseline comparison (benchmarking)of "Sites of similar size" is provided under the visitors tag above the Google Analytics. You can choose a similar type of site with your site as a baseline data comparison, this is a good reference, because through comparison can be more specific to your site in the same type of site advantages and disadvantages, to optimize the direction of the site. GA with its powerful data platform can provide us with the baseline, but perhaps for the above example will appear powerless, this time we need to rationally choose a suitable baseline, such as my blog is currently the category and content are relatively small, Then I might define my site's lost user ratio should be controlled below 1%, but if there is a more complex web site for applications and content, then the baseline will obviously need to be set a little higher. Once a certain period of time data across the baseline, it is necessary to pay attention to the site's information architecture is not in the confusion, it is not to do some sorting and optimization.

In short, a good information architecture can help users find the information they need more easily, thereby effectively enhancing the user experience of the site, so try to optimize the information architecture of your site. If you have a better way to effectively test the quality of the information structure of the website, or to be able to clearly analyze the details of the Web site information Architecture defects, I hope to communicate with me, I look forward to the site analysis methods in the optimization of the site information Architecture more applications.