Attention Spans: A Theory

Attention spans aren’t simply on or off. There are phases that you need to move through by earning your audience’s trust.

Over the last year and a half I've participated in something like thirty to forty remote user tests. And all of those hours of testing have led me to the conclusion that something is very, very wrong.

As product designers and engineers, we do not understand the average user's attention span.

On a typical remote user test we’ll have a dedicated user tester, the appropriate product manager, and a product designer (that’d be me) all share a screen with an existing customer. This customer will be sitting down at their place of business and will be asked to work through a few simple tasks. Minute for minute, it's one of the best ways to test our assumptions, how well we executed on designing for those assumptions, and how well those assumptions stand up against the actual needs of the customer. I always leave with a bittersweet sense of having gained a number of valuable observations (based on what I’ve seen), but remarkably few meaningful inferences (that is, what I think it all means).

One of the reasons why making meaningful inferences is so difficult is that there are so many possible explanations for why someone might do something. How the customer uses his or her attention span is a critical part of untangling this mess. It’s also a part that seems to be consistently misunderstood or ignored.

The primary problem: We think attention is either on or off

Attention works differently depending on how much of it you give. How often do you read an article while thinking about something else, then have to go back and reread the last few paragraphs? How often have you been listening to someone else talking, while thinking about something else, and missed the main point of the conversation? It happens. I do it all the time. I'm not judging (unless you’re doing it to me — that’s not cool).

A very common observation during a user test is when the participant misses the instructions that would have helped them complete the requested task. The easy inference is that people simply choose not to read, or not to read thoroughly. But this conclusion makes the error of confusing an attention span with a conscious choice. Maybe they didn't notice the text because they weren't paying attention instead of noticing it and choosing not to read it.

Now, this could easily just be a self-serving attitude on the part of a designer who wants to defend an interface that someone didn't use as expected. You might be right (don’t tell my boss), but there's also an unanswered question here about how attention spans are gained and lost.

Phases of an attention span

Whenever I have the chance to sit next to a customer (gasp! in person?) and watch them use the software as they normally would, I notice a much different behavior than I do in a remote user test. They skim through more screens, they bounce back and forth between screens, and they sit and contemplate other screens. From what I can tell they seem to be moving through a spectrum of their attention and changing their behavior along the way.

Here are the the three big phases I'm noticing:

1) Passive gaze (minimal attention for a second or two)
2) Focused gaze (moderate attention for 5 to 10 seconds)
3) Choosing to act (full attention until task is complete)

Passive gaze

Immediately after seeing an interface, it seems as though the user is looking for shapes and colors and taking guesses at what the overall layout might be trying to communicate to them. Is this a form? Is this a stream of activity? Is this a chart? Is this what I thought it would be?

If the user was expecting to create something through a form and doesn't immediately see something that looks like a form they will abandon their goal. They can't cope with the cognitive dissonance and will most likely try to escape by hitting the back button or searching the navigation systems for something more appropriate. No tutorials, microcopy, or help videos will save them from this level of expectation failure.

To earn your user's passive gaze you need to a few things:

1) Carefully set expectations in previous interfaces with clear buttons, links, or calls-to-action that prepare them for what's to come
2) Prominently reiterate the expectations set in the previous interfaces
3) Consolidate the look and feel of interfaces with similar purposes so the user can immediately map visual components to their expectations

Focused gaze

If the user feels as though they're in the right place and have access to the right tools and information, then they’ll generally commit to the next phase of attention. A focused gaze seems to be about the user establishing a timeline and guessing at the effort that’s about to be involved with completing their task. They want to see the light at the end of the tunnel and to feel as though the effort is going to be worth the expected outcome.

So if the user is looking at an article, then they're skimming headlines and a few subheadlines; if they’re in a social media environment, then they're glancing at the first few pictures; if they’re looking at forms, then they're scanning through the first few labels; and inside data-heavy screens, they're trying to find the tools that will help them control what they see (i.e., filter and sorting features). None of this behavior is specifically learning how to do anything. It's almost entirely about getting a sense of the time and effort they will need to spend in order to make some kind of progress. Again, any tutorials, microcopy, and help videos might have very little impact during this phase.

To earn your user's focused gaze you need to a few things:

1) Provide short and precise phrases in headlines, form labels, and tool titles to help people make quick decisions about the purpose of the screen
2) Keep similar things close to each other so people can group areas of the screen in their mind
3) Avoid introducing secondary goals until users have a strong sense of the primary goal

Choosing to act

If the interface has met the user’s early expectations and given them a way to judge the time and effort involved, then they can confidently choose to take action. Only after they have made this choice can I (the tester-observer) finally start to make better inferences based on my observations. The user’s attention span is finally “on” and so the behavior I’m observing has much more meaningful implications.

After a user has started consciously investing their time and energy, they're more likely to look for help when confused. This is where tutorials, microcopy, and help videos can be used effectively. When a question is confusing on a form someone has chosen to complete, they'll take the time to read the microcopy. When trying to understand a complex chart, they take the time to read the legend, axis, and inline help text. After deciding to build a landing page, they can then take the time to watch the video tutorial.

To keep your user's full attention once they’ve chosen to act, you need to a few things:

1) Keep help content in close proximity to where people will get confused so they don’t have to look for it and lose their flow
2) Design help content to be easily ignored if need be, so that it doesn’t disturb the people who don’t need it
3) Keep help content appropriately paired to the time and energy expectations of the user

Do I still have your attention?

I want to continue exploring how people invest their attention, and I’m sure many of those who are reading this have thoughts and insights to share. If you have and links to articles that might help give more context, please suggest them below. Thanks for reading!