Recently, we conducted a large-scale survey to understand how people make important decisions based on online information. The survey replicated and expanded a study conducted by Xerox PARC researchers in 1997.

Compared to the original study, we found that critical internet usage today involves more activities directed towards gaining an understanding of a topic than 22 years ago. We also noted that a fair percentage of the critical information was acquired passively, meaning that the user did not set out to find that information.

In this article, we focus on passive information acquisition and how it differs from active information acquisition (where users had a prior intent to locate certain information). Our study revealed that over 22 years, passive information acquisition increased from 4% to 14% of all incidents leading to significant decisions or actions.

We identified two types of activities associated with passive information acquisition:

  • Explore: information was found accidentally while browsing
  • Notified: information was included in an email, text message, or push notification

Active vs. Passive Information Acquisition

We found 5 differences between active and passive acquisition of information involved in critical decisions. Compared to active information acquisition, we observed that instances of passive information acquisition had:

  • More social interaction
  • Different user expectations
  • Fewer content types covered
  • A stronger influence of ads
  • Fewer devices used

More Social Interaction During Passive Information Acquisition

We observed that people were significantly more likely to interact with others when they gathered critical information passively than when they engaged in active information acquisition (p <.01).  In other words, when people discovered important information without looking for it, they were more likely to want to interact with others.

Frequency of Accompanying Social-Interaction Behaviors
People were almost twice as likely to interact with others during passive information acquisition than during active information acquisition.

As described in our methodology article, our analysis of the critical-incident data identified 5 types of social interaction:

  • Collaborate: work together with others to reach a decision
  • Inquire: ask someone else for additional information
  • Informed: get information from others (without asking for it)
  • Share: share info with others
  • Execute: accomplish an action or goal through social interaction (e.g., call to make a purchase)
Accompanying Social-Interaction Behavior Types
During passive acquisition, people were much more likely to share information with others than during active information acquisition.

For passive information acquisition, Share was the main type of accompanying social interaction. People shared significantly more during passive information gathering than during active information gathering (21% vs. 2%).

For instance, a respondent browsed the news, saw a report of tornados in Indiana, and then she shared it with friends to warn them. Another user praised the convenient Share functionality in Yelp, as her friends could quickly suggest where to eat in Miami.

The Sharing Feature of Yelp
Yelp mobile app: People could easily share restaurant information with others.

In contrast, during active information seeking, the most common social interaction (15%) was Inquire — usually to get more information in order to finalize the decision. For example, a respondent comparing different mortgage options “called the bank with follow-up questions and settled on the loan and grant option that fit [her] best.”

The BestBuy site allowed immediate communication between shop representatives and buyers, through a Help button available on product pages. Specific questions related to the products could be directly answered by specialists.

Best Buy_Help Center
The BestBuy site allowed immediate communication between users and shop specialists. This feature supports Inquire-type of social interaction.

More Context, More Customization Expected for Passive Information Acquisition

We asked people if there was anything that they would like to change on the sites or apps they used for passive information acquisition. Using thematic analysis, we identified the top issues people mentioned and compared the expectations for active versus passive information acquisition.

Related and Credible Information Were Valued for Passive Information Acquisition

Though content issues were the most reported category for both active and passive critical incidents, users valued different aspects of content during the two types of activities.

Active information seekers wanted detailed information presented in multiple formats (e.g., text and video) and capturing various perspectives (both content providers’ and users’). They also wanted the information to be clear and easy to consume. For example, a user who researched and bought a washer-dryer combo wanted images of all sides of the product. Another respondent who learned a game rule wanted the information to be “more direct and to the point.”

But during passive information acquisition, the main issue was lack of context because users often came across the limited-scope information accidentally and in isolation — for example, they encountered a piece of news or a single product shown by itself, without surrounding context. They wanted to learn more about it or view more options. For instance, a user who had browsed an article regarding the mistreatment of animals and decided to become vegan wished that the article had “links to more information.”  Another user who was reading about CBD oil wanted to be able to easily order the products mentioned in the article.

Thus, providing contextual information can help users act upon the content they discover accidentally. At a minimum, provide links to related content, though more extensive contextual features like comparison tables or polyhierarchical information architecture can also help, depending on the nature of your content and product space.

Users also emphasized the importance of content credibility and transparency while receiving critical information passively. One user reported reading an article that convinced her against voting for a certain politician; she wanted the article to “cite sources clearly.” Another user who saw an article about the Rock ‘n Play Sleeper recall reported, “I wish there was less sensational news or click bait.”

Though many people make decisions based on information they passively receive, it doesn’t mean that it is okay to cheat them. Link-bait and inflammatory headlines may increase the conversion rates in the short term, but in the long run, such tactics will diminish the users’ trust in your sites.

Customization Valued for Passive Information Acquisition

Notifications have shown a profound impact on user behaviors. Users mentioned the need to personalize and customize the experience, instead of being overwhelmed by tons of notifications that didn’t meet their interests. For instance, a person who browsed news about a baby-product recall and was concerned about it reported, “I would like to be able to follow news about that product.” Another user wrote, “I want only important news and weather to send me push notifications.”

(In general, the ability to be informed about future events and other forms of follow-up information is extraordinarily poorly provided on today’s internet. This means that doing so well is a major opportunity for individual sites to stand out and establish direct connections with customers that bypass search engines and social-media sites.)

A Narrower Content Scope for Passive Information Acquisition

If we take a closer look at the categories of information that people passively acquired, we could Interestingly find that they were less diverse, compared to the information acquired actively: Product Info & Purchase, News & Politics, and People covered 81% of all passive info acquisition. This finding may be due in part to the smaller number of passive info-acquisition incidents in our study (71 responses vs. 427 responses in the category of active information seeking).

However, these three categories are also commonly explored during browsing activities. When users scroll through social media or read the news, they may not have a specific goal in mind, so they may be open to switch to a different task if they stumble upon unexpected information passively.

Product Info & Purchase was the most-reported content category for both active and passive info acquisition. People were more likely to find critical Health information actively (p < 0.001), which makes sense — most people actively seek out answers to health questions or problems. On the other hand, higher proportions of passive activities fell under News & Politics (p < 0.001), and People categories (p < 0.001), compared with the respective proportions of active incidents.

Top Content Categories
Similar proportions of critical information in the Product Info & Purchase were gathered both actively and passively. On the other hand, critical information related to Health was more likely to be gathered actively, whereas important content related to News & Politics and People were more likely to be gathered passively. (Note: We only show the 4 content types which were either most reported or showed statistically significant differences. There were 13 content categories in total. For the 9 rest categories, there were either no significant differences or the number of passive incidents was too small for a meaningful comparison.)

Paid Content Can Have a Strong Influence on Decision Making in Passive Information Acquisition

We’ve been reporting for years that users instinctively avoid ads; this behavior results in phenomena such as banner blindness. However, while people find most ads annoying and unhelpful, that doesn’t mean ads are always ineffective. When an ad is correctly targeted, relevant, and carefully presented, it can have the desired effect on its users.

Paid content, including ads (31%) and promotional emails or notifications (6%), accounted for 37% of the critical passive information acquisition that we observed. In contrast, only 1 out of 427 respondents mentioned being influenced by ads during active information gathering.  

What’s the reason for this difference? When users are engaged in passive information acquisition, they are less goal-driven and thus more susceptible to have their attention captured by a variety of content — both organic and paid. On the other hand, in many cases when people are pursuing a goal, online ads are intrusive and unrelated to that goal.  

The sources that people reported for critical paid content support these assumptions: social-networking sites (e.g., Twitter, Facebook) and browsing-oriented services (e.g., Hulu, Reddit).

  • “I saw an ad on Facebook for discounts on running shoes, and I ended up buying the shoes.”
  • “Buying my current phone, Google Pixel 2 XL. Kept seeing commercials on Hulu about it. Usually avoid commercials like the plague.”

Thus, it is important to align your ads with the potential goals and interests of your users, but try not to disrupt them. When we asked if users had any suggestions for the sites or apps that they used to find the critical information, users performing passive information acquisition didn’t complain about ads too much. But respondents performing active information acquisition commonly mentioned being frustrated by disruptive ads placed in the middle of content. (Not surprisingly, since, in another study, we identified this type of ad as one of the most disliked ads for the desktop.)

More Mobile Use During Passive Information Acquisitions

Compared to active information seeking, people were less likely to use large devices like desktops and laptops when performing passive information acquisition. The difference was marginally significant (p = 0.06).

Also, when critical information was acquired passively, people were less likely to use multiple devices for one task. 22% of people who performed active information seeking reported using more than one type of device, but only 10% did so for the passive information acquisition (the difference was statistically significant, p < 0.05).

Device
When passively acquiring critical information, people were less likely to use large devices like laptops and desktop computers.
Percentage of One vs. More Devices
Fewer devices were involved in passive information acquisition. When actively acquiring info, people were more likely to use multiple devices during the same task.

This result seems straightforward. The passive information acquisition is unplanned and accidental and often occurs when people are killing time on their devices — mostly on their smartphones. The other devices may not be available to them or the information received through a notification or an email may create a sense of urgency and encourage them to make decisions quickly.

Imagine it’s 8 pm, and you are browsing social media casually after dinner. An ad on social media informs you that there is a discount on a skincare product you love. You tap the ad and see a 50% discount available for a short time. So you purchase the product right away, on your phone. This is a scenario reported by several participants in the survey. The action is driven by a sense of urgency and scarcity, and people are less likely to switch devices during these sessions — they are afraid of “losing” the deal.

Conclusion

Our research shows that an increasing number of people find critical information accidentally, without specifically looking for it. Emails and push notifications, as well as content discovered through browsing, can serve as a basis for decision making — provided that the content is sound, self-explanatory, contains sufficient details, and is tailored to the users’ general interest.

Designers have the responsibility of making their content discoverable, by creating strong information scent for that content. Good, accurate summaries for articles, clear calls to action that take people into the right workflow, seamless transitions across different channels (such as from email to website) can all ensure that users fully benefit from relevant information encountered accidentally across the web.

We’ve seen that the content that is perceived as relevant (including ads) is likely to be appreciated by users. But what is relevant can only be determined if we know our users and understand their interests and behaviors. On the one hand, this means exploring our audience as a group — conducting user research in order to figure out general goals and tasks. But it also means knowing people at the individual level, by taking into account their history and their previous interactions — a feat that becomes more and more attainable as machine-learning technologies increase in adoption and accessibility.