Summary: In interface design favor direct access to the user’s preferred item instead of forcing users to go through your content in a serial order.
If you happened to be around in the 90s, when the web was invented, you may remember that "hypertext” was the buzzword at the time. In fact, “HTML” itself stands for “Hypertext Markup Language.” Hypertext made the web work as an interconnected media form: text that contains links (hyperlinks) to additional content that can be immediately accessed.
The hypertext and hyperlink exemplify the direct-access paradigm and are a significant improvement over the more traditional, book-based model of sequential access.
(Direct access can also be called random access, because it allows equally easy and fast access to any randomly selected destination. Somewhat like traveling by a Star Trek transporter instead of driving along the freeway and passing the exits one at a time, which is what you get with sequential access.)
In a normal, physical book, the reader is supposed to read pages one by one, in the order in which they are provided by the author. For most books (fiction, at least), it makes little sense for the reader to turn directly page 256 and start reading there. Unless, of course, that is where the reader left off in their last reading session. Getting to page 256 in a 500-pages book poses a bit of a challenge, as we well know it, and each of us have their preferred method of dealing with it (be it a bookmark, a dog ear, or our own memory).
Tables of contents try to alleviate a book’s sequential-access problem by telling people what content is going to be found in the book and at which page. The user still has the problem of turning to the desired page number, but at least he doesn’t need to bother with parsing the content and deciding whether he’s found what he is looking for.
By definition, however, the web embraces direct access. Thus, it is disappointing to see sequential-access designs becoming increasingly popular nowadays.
Costs and Benefits of Sequential Access
But why is sequential access so bad? Simply because it forces the user to work harder than she needs to: she has to process all the content that sequentially precedes the piece of information that she is interested in. Thus, sequential access increases interaction cost.
Sequential access has two potential benefits:
Progressing linearly through an information space can be accomplished through particularly simple navigation controls: basically a “give me more” button. However, designs like infinite scrolling often hurt users more than they help. You ought to design navigation controls that allow users more freedom without being overly complicated.
If you know that users have been through the earlier steps in a sequence, you can build on that knowledge in explaining the next step. In practice, of course, users often scan verbose web pages and miss much of the information. So you can’t truly rely on users reading (much less understanding) all the earlier exposition, even if they have passed through it.
The benefits of sequential access are more hoped-for than they are real on most practical websites. In contrast, the costs are very real and are incurred every time.
Examples of Sequential Access in User Interfaces
Let’s take a look at a few examples of sequential access in modern interfaces.
The carousel has always been a popular way to stick content on the front page without taking up too much space and has seen a resurgence with the advent of the iPad. (Original iPad designs were fascinated by the etched screen aesthetic and wanted to control the layout in the tiniest detail. As a result they often forewent vertical scrolling in favor of a card or carousel-like design.)
Carousels have many advantages, but one big disadvantage is that they are based on sequential access: users must go through all the items in the carousel one by one in order to get to the last one. This interaction is inefficient and provides little information scent: users generally have no information about what comes next. Although carousels may solve content-priority quarrels within the organization, they slow users down (at least in their more traditional incarnations) .
How can you make carousels more direct-access like? If you cannot avoid them altogether, provide links to the stories in the carousels to let people select them in any order or, at least, present more than one item at once.
Also, remember that carousels are ok only for short lists: users should be able to get to the last item in the list in 3–4 steps. Search results or long lists never belong in carousels; as one of our users put it, “I don’t know what item 20 is, but I know that I will never find out.”
Even more than books, videos are the sequential-access medium par excellence: users must patiently watch a lot of video footage before getting to a piece of content that is relevant or interesting to them. That is why videos by themselves are not an ideal medium for instructional or informational content; although they can work great in conjunction with text, if they are the only method available to users, they are terribly inefficient.
How can we fix them? Not all material needs to be in video format. If you provide a video, make sure that you also provide, if not a transcription, at least a detailed text summary that allows people to quickly scan the information for relevant details.
With the advent of responsive web design, uncommonly long pages proliferate not only on mobile, but also on the desktop. A long page that contains a variety of content forces the user to scroll down with the hope that they will find something relevant. Yes, users scroll, but only if tempted by the promise of relevant content. If the page information is made of different, loosely related pieces of information, users have no way of knowing whether they must scroll for more or they should stop. They often err on the side of minimizing effort and stop before reaching a relevant piece of information.
How to fix the issue? Avoid excessively long pages altogether. If you cannot, at least provide a mini-IA: a linked page table of contents at the top of the page or use accordions to do so (especially on mobile). The mini-IA (whether from accordions or jump links) will tell people what to expect on the page, it will allow them to form a mental model of the page, and will also facilitate direct access.
Screen readers and keyboard-only navigation exemplify another one of the pitfalls of sequential access. These tools scan all the links on a page in a sequential manner. If the link of interest is somewhere in the middle of the link list, it may take forever to get to it and be able to select it.
When iPad magazines first came around, they used to follow the physical-magazine mental model and eliminate all direct access and hyperlinks. Stories were referenced on the cover or in the table of contents, but they were not linked to. Luckily most publishers eventually realized that a lack of hyperlinks was a tremendous downside because it forced people to browse through stories as if they were using a paper version.
How can we fix the issue? Use hyperlinks.
Selection from a Long List on Mobile
On mobile devices we often encounter designs that favor selection versus typing; these designs are based on the assumption that typing is difficult with a small touchscreen keyboard. As a result, users are sometimes forced to select an alternative from a long list of items — for instance, a list of years or countries. Good idea? Yes and no. It is indeed generally easier to select than to type, but not if you have to scroll a lot to find the item of interest. (On desktops, often the names are all visible and it is ok to let people see them at once and select one from them.)
How can we fix the issue? Allow people to type 1–2 letters and offer suggestions based on those. Even though typing long words or phrases is painful, it’s not so bad just to enter the first character(s). People can still select, but now from a list that has been considerably narrowed down to only a few alternatives. (Like alphabetical sorting, this solution works if the names of the items in the list are known to the users, as it is often the case with countries, brands, and car makes. )
When Is Sequential Access Appropriate?
Sequential access is the method of choice if you expect users to access all the content in a prescribed order. It forces users to accept your curated contribution and assumes that most users will be willing to do so. For works of fiction, many articles, or entertainment videos that assumption is accurate. If, on the other hand, people are likely to be unequally interested in all the content that you offer, use a direct access method to let them reach their goal faster.