In most markets, products and services compete for the consumer’s money. On the internet, however, the coin of the realm is time, not money—websites and other online services fight for the attention of visitors. So understanding when, how, and for how long we allocate our scarce time for activities is crucial to online success.
A recent research paper offers insights that carry unexpected implications for advertisers or anyone else trying to capture that attention. The Empirical Economics of Online Attention was written by Andre Boik, University of California at Davis; Shane Greenstein, Harvard Business School; and Jeffrey Prince, Indiana University.
The study notes, for example:
- We devote a more or less fixed amount of time online each week. We don’t allocate more time when something new comes along; instead, we shift our fixed amount of attention to new sites at the expense of old sites.
- Our online visits often come in short bursts rather than extended leisurely strolls through cyberspace.
- People with higher incomes spend less time online than those making less.
In other words, consumers behave online in a much different way than they do in markets in the physical world, where they will carve out more time for shopping and for sniffing out the best bargains.
"What we observe suggests that households have fixed amounts of time to give to sites..."
“Taken together,” the researchers conclude in their paper, “our results add up to a striking model of competition for attention.”
HOW WE BROWSE
For their study, the researchers analyzed browsing activity on the primary home computer of more than 40,000 US households in 2008 and more than 30,000 in 2013, studying how much attention was allocated, where that attention was allocated, and how that attention was allocated. Clickstream data was provided by ComScore, an audience measurement firm.
In the development of the internet, 2008 and 2013 were landmark years. Netflix and Hulu began offering streaming services in 2008, and YouTube videos were all over the Web. Smartphones and tablets provided new ways to access the internet. With such expanding amounts of appealing content, it seemed feasible to Boik, Greenstein, and Prince that people would have visited more sites during that period, and spent more time on them as well.
But that’s not what happened, the researchers discovered. Instead, both numbers remained remarkably stable over the five years. There was an overall shift in the types of sites visited, however, from chat rooms and news to social media and video.
“We were taken aback,” admits Greenstein, the MBA Class of 1957 Professor of Business Administration.
The stability of the time spent on the internet might be a bit more understandable if you take into account what anthropologists have discovered—that internet use inside most homes is “bursty” and “plastic.” In other words, we use little moments of free time, what the researchers term slots, to go online—the 10 minutes between dinner and kids’ homework, or the five minutes before your favorite TV show starts. The researchers confirmed that observation.
“There’s an underlying inflexibility here,” Greenstein says. “What we observe suggests that households have fixed amounts of time to give to sites, and that going online is a ‘filler’ activity.”
Researchers also found that households with annual incomes of $25,000 to $35,000 spent an average of 92 minutes more time online than households making $100,000 or more. (The average daily amount of time spent online in a US household was two hours.)
While internet penetration in US households has been high for years, that number still increased, reported Boik, Greenstein, and Prince. In 2007, nearly 62 million households had access to broadband; in 2013, that number has risen to 73 million, with the majority of adopters falling into the lower-income category.
ADVERTISERS TAKE NOTICE
For web-based businesses looking to attract users, the implications have some interesting strategy considerations, although this is not a large focus of the study.
“To reach people, you can’t do what you usually do, which is have a sale,” Greenstein observes. “You can sell your products for less, but you still have to get people to your site. There is no price that will cause people to allocate between different choices, which makes this unlike any other market activity.”
The bursty nature of online attention also helps define what is required for effective advertising or promotion. “If you have a concept to communicate that is 20 minutes long, you will be competing with other instances of 20-minute content,” Greenstein says. “Don’t expect households to take two 10-minute slots and put them together for you. That’s not going to happen. Similarly, if you have three minutes of content, you’re competing with other three-minute content sites; you shouldn’t consider households with six-minute slots that they’ll divide. That’s a stark implication of what we found.”
It would seem that one bright spot favoring advertisers is demographics, since internet adoption (at a solid 99 percent penetration) skews to high income, high education households. However, those users don’t spend as much time online as lower educated, lower income households, making the window for catching the attention of the affluent even smaller.
“From an advertiser’s perspective, those two statistics work against, rather than reinforce one another,” Greenstein says. “With that said, when these findings are projected to US demographics, it is still slightly more likely that an advertiser will capture the attention of a higher income household than a lower one.”
As internet adoption became more ubiquitous, the average online encounter shifted to someone of lower income and longer sessions. “We expect that trend to continue,” says Greenstein.
The researchers note that the typical household spends more than half its time at two sites. Targeting within those two sites is one approach to reaching users. The typical household also spends more than a third at an enormous variety of specialized sites. Targeting specialized readers at the entire site is another approach.
FUTURE STUDIES
In addition to adding more recent data to the research, Greenstein hopes to look at information captured on other devices, such as smart phones and tablets.
Another question to be answered: Where are people visiting when they are not on the top sites that most everyone looks at? With some 4 million URLs in the dataset, Greenstein wants to explore consumer behavior on sites outside of the Top 3, which in 2013 were Facebook, YouTube, and Google. (The Top 3 in 2008 were MySpace, Yahoo, and YahooMessenger).
“The top three sites for most households is the same—but when you get to four or five, they could be anything. Household preference variance is something we’re very curious to look at more closely, because those sites account for a significant percentage of online attention.”
Getting a clearer window into how households allocate what little available attention they have could have implications for other markets, Greenstein adds, such as television and radio. While their study doesn’t focus on either medium, a similar dynamic could apply, with an increase in the supply of television (and ways to view it) causing a shift to new content and perhaps a small drop in how it’s viewed, but no real change in the allocation of time or watching behavior.
Whatever the format, gaining a richer understanding of how attention is allocated can only become a more pressing question for content producers and advertisers. And that understanding is in its early stages of development, leaving many open questions to explore.
“Online advertising is still a work in progress, with considerable effort being invested in improving its cost effectiveness,” says Greenstein. “Those efforts—and lessons learned—are relevant across many formats, not just PCs. There is so much more to come.”
Related Reading:
Dear Internet: You Are Extraordinary But Not Exceptional
Advertising Symbiosis: The Key to Viral Videos
Encouraging Niche Content in an Ad-Driven World