Perceived cost-benefit and (online) behavior

Often I hear these broad blanket statements on what people do or don’t do online: “people don’t read,” “people don’t scroll,” “people don’t wait.” It’s usually from those who have just enough knowledge of usability to be dangerous, but sometimes, alas, I’ve heard usability people uttering them. When I hear statements like this I’ll usually slowly count to 10 to regain self-control and then with the calmest tone of voice I can fake I’ll say: “Well, let’s talk about it.”

So, let’s talk about it.

Do people read online? – Of course people read online. Don’t you go online to get your news? Aren’t you reading this blog post online? You read content online all the times. But even a more specific statement such as “people don’t read instructions online” is incorrect.

First of all, not many people read instructions, whether online or in print. This is why when instructions take more than one page, smart technical writers have created these handy one-page Quick Start guides. Most people read manuals when they are in trouble, not to learn how to use something. (I am one of those few peple who sometimes read the entire manual before starting to use the thing; but then, I used to read the dictionary for fun, so you can see where a lot of my psychological problems come from.)

For example, when you observe people filling out forms you’ll see that the most common strategy is first trying to complete the form without using text instructions (but using other important written pieces of information like field labels, formatting tips, and examples); only in case of uncertainty or failure, people go back to read written instructions.
However, there are plenty of exceptions. I’ve frequently seen people reading with great attention seemingly irrelevant blocks of text, even lengthy ones, both online and in print.

There are many reasons why people read or not read text. But there is a simple way of thinking about it that is more sophisticated and useful than a simplistic, black an white statement like “people don’t read.”

Cost-benefit assessments drive behavior – People make decisions continuously on what to do and what not to do. When a person is in front of a form, he or she can decide to read any piece of content, read only part of it, scan for relevant information, or start filling out information right away. We can think of the decision at each point in time as determined by an analysis of the perceived cost-benefit based on the person’s goals and constraints. If the perceived effort in performing an action is higher than the perceived benefit, the person will not perform the action; and vice versa.

Let’s consider for example a person filling out a simple form. If the label of the first field reads “Enter your name” she would probably skip long text instructions and glance at the example showing the order of first and last name to understand how to fill out the form. The apparent simplicity of the action and the low probability of error doesn’t justify the effort required to read the instructions.

But let’s assume that the person encounters something less familiar and more complicated. If the strategy of relying only on labels and examples doesn’t work, she would probably go back to the written instructions to understand what to do. If the perceived cost of making a mistake or being unable to complete the form is higher than the perceived cost of reading the instructions, the person will read the instructions.

The way the text is presented also influences the perceived cost of reading it. A bulletted list of short sentences is better that a long continuous paragraph because the perceived cost of reading it is smaller.

Influencing the perceived cost of reading – There are many strategies to decrease the perceived cost of reading text. For example, highlighting keywords in bold supports rapid scanning, which is a more efficient alternative to word-by-word reading. Font type and size, spacing, and other typographic factors all influence the perceived effort of reading text. (Incidentally, they also influence the perceived importance and relevance of each piece of text, which in turn influence the likelihood that someone would read it.)

This means that designers can influence the likelihood that people will read or not read text; they control the visual characteristics that determine the perceived effort required to read it.

The quality of content, of course, is what eventually determines whether people will read or not and how much people will read. Is it the right content? Is it relevant to the reader? Is it concise and clear as it could be (in other words, is it respectful of the reader)?

Trust and risk – Non-design factors such as trust or risk also enter in the cost-benefit equations. For example, I am currently testing a print form for an important financial transaction that is associated with a certain level of concern and preoccupation. I’ve never seen so many people reading the small print and spending so much time trying to understand the instructions. In this case, reading is perceived as less costly than making a mistake in filling out the form and risking errors and delay in the processing of their request.

The influence of perceived risk might also explain why people seem more inclined to read instructional text in print than online. Experience teaches that if one makes an error on an online form, one will probably receive an error message; if one skips a mandatory field, he or she will be unable to continue to the next page. It’s like somebody is looking behind our shoulders and checking our work at each page. With a printed form, we are on our own. If we make a mistake, nobody will point it out to us. So we’d better be careful.

The probability that people would read the “fine print” seems to be inversely correlated with the trust they have in the company they are dealing with. What I typically hear is something like: “I don’t need to read all this. I’ve always trusted them,” versus “I don’t know them very well, I’d better read what I am signing for.”

Putting all together – One thing that I like about this way of conceptualizing behavior is that it recognizes that people’s behavior is intelligent and driven by efficiency. If people don’t read my instructions is not because they are stupid or lazy, but because I am: I’ve written or visually presented instructions in a way that conveys an effort greater than the benefit to read it. I’ve not spent time understanding people’s goals, motivations, and constraints. This also mean that I can do something to fix the problem, by changing the perceived cost-benefit balance in my favor.

The same reasoning works for any human behavior. Why people threatened by hurricane Rita and who have been urged to evacuate decided to stay? If you think about the perceived cost/effort of the two alternatives (“should I stay or should I go?”), together with the trust or mistrust in the authorities that ordered the evacuation, the perceived risk of being killed or injured by the hurricane based on previous experience, and the observation of the behavior of other people they trust, you can put together the cost-benefit analysis that supports people’s decision to stay.

Follow by Email

6 thoughts on “Perceived cost-benefit and (online) behavior

  1. I’ve been experiencing something lately where people think the design is broken and say “Make it bold”, “Make it bigger” when sometimes a usability problem could be solved with some simple content editing.

    Editing a piece of text to make it shorter and more concise can do a lot for scan-ability. Or moving the most important key words or sentences to the beginning. So often, people critiqueing a web page will look at the size & weight of the font and not even think about what the content is saying. They say “We’ll figure out the content later, let’s just make it bold so it stands out.”

    Well… if EVERYTHING on the page were bold, then bold text would have no emphasis at all! What will you do when you really, really need it? Bolding everything on the page would break any sense of hierarchy.

    Good design is about communicating a message, not just picking colors & fonts, so let’s think hard about WHAT we want to say, not just how we want to present it.

  2. You are right, Laurie. The content aspect of design is as, if not more, important than the visual aspect of design. If you don’t have the right content to “help”, or if you have too much content, or if the terminology is confusing, you are in trouble.

    But from the point of view of the person using the application, it doesn’t matter if it’s content or information design. People will say that they don’t understand or they don’t see: it might be that the right keyword is missing, or that the instructions are hard to understand, or too verbose; or it might be that the content is there but it’s hidden by other visual elements, or it’s in the wrong location and they cannot find it. Either way, they will have problems in using the interface.

    It’s interesting that when you talk to writers, they have the same complaints that you have from their point of view. They will say that everybody want to fix things with content. If the design or the information architecture don’t work, they often are asked to “improve usability with content,” because content is perceived as cheaper to fix. Content can be patched at the end after coding is completed. If you think about it, it’s quite ironic: sometimes the same people who say that users don’t read also ask to fix usability issues with content.

    Designs is effective when everything works together: The logical structure of the interface, the visual structure of the interface, and the content (and even the objectives). To me, this is design: a seamless, useful, usable, and desirable interface. Content, IA, and information design are different skills, but from the point of view of who is using the interface, they are artificial distinctions. We need to find a way to work together to create something that works.

    A final thought: when people first encounter an interface (a web page, the first page of a printed form, or any other UI) they quickly create a first impression that will determine their approach to the interface (see Christine Perfetti articles on the 5-sec test). This first impression is created by a mix of page structure, visual organization, and visible keywords and is very important. Again, it works effectively only when all the pieces that people perceive are integrate effectively.

  3. I realize I’m one of the most verbose writers in the Blogosphere and probably have no business commenting on this subject. But, quite plainly, good content gets read. Hopefully. These generalizations about what people will or won’t read often has to do with not willing to read what they’ve already read elsewhere. Good “flow”–something that reads well, breaks up a bit, says what it needs to say uniquely and concisely–will appeal to me every time.

  4. Just a thought,

    I suppose it is a matter of not only what is utilised to scour for information but who is doing the scouring. For instance, the mouse enables one to exit an experience with a single click. Whether a person decides to utilise this in a manner not unlike the couch potato’s remote control is determined by the collective character prior to and in the course of this experience.

    For instance, a person may approach the net experience with a desire to know. However, this desire to know, in the face of a culture which is increasingly designed to appeal to the senses of an individual, may be defined by it. Thus, bangs and whistles may direct the viewer to what is worthy of being known. The net experience, by being of the same sensorily stimulating school as Hollywood movies, adverts, and now including even nature channels, which in turn, is complemented by the Type A personality of Capitalism; the educational system increasingly being geared to adapting to its demands; the intellectual deficiencies pandemicised by the division, specialisation of labour; the abbreviation and Simpsonisation of language and thought; all serves to increase an ‘individual’s’ propensities to be appealed to by the more sensational rather than the cogitational.

    Additionally, the ‘blog’ experience serves to qualify an opinion simply by enabling anyone to have a digital soapbox. This can lead to the user confusing the ability to express an opinion on a prominent podium to being a validation of the quality of the opinion. Thus, this can lead to a polarisation towards concordant views or those that are most simple to understand and enjoying mass favour. From this point on, the ‘individual’ starts to seek out information selectively and becomes victim to the abbreviated and sensational.

    Consideration, in the light of the above, is thus subject to necessity rather than individual propensity.

  5. There is definitely a social influence in what we consider worthwhile reading or watching. Doesn’t it start when you are a child and you need to watch the same TV shows that your friends watch, so you can talk with them about it and feel you are part of the in-group?
    We are social animals, no doubt about it.

    So, part is your motivations and the goal you are trying to achieve; part is your personality; part is the medium you are using; part is the habit you have formed. And finally, what you have learned to perceive as relevant or not relevant in your social environment.

    Still, I think, in designing content there is a way to influence at least partially the way people behave. I’ve seen it so many times in my usability tests: there is some invisible text almost nobody sees, then you tweak the verbiage or change the visual appearance even slightly, and suddenly most people now see it.

Comments are closed.