How e-commerce websites manipulate you, even if you don’t want to buy things

When potential customers visit the ThredUp online resale store, regular messages on the screen tell them how much other site users save.

“Alexandra from Anaheim has just saved $222 on her order,” says one message next to a picture of a bright, multicolored dress. It is a popular method on shopping websites, designed to capitalize on people’s willingness to fit in with others and generate a “fear of missing out.” But “Alexandra from Anaheim” did not purchase the dress. She doesn’t exist. Instead, code from the website pulled combinations from a pre-programmed list of names, locations and items and presented them as actual recent purchases.

False messages are an example of “dark patterns,” devious online techniques that manipulate users to do things they might not otherwise choose. They are the digital version of time-worn tactics used to influence consumer behaviour, such as impulse purchases placed near cash registers or bait-and-switch ads for used cars.

Sometimes the techniques are obviously deceptive, as with ThredUp, but often they walk a fine line between manipulation and persuasion: think of the brightly colored button that promotes you to agree to a service, while the opt-out connection is concealed in a drop-down menu.

Web designers and customers have been highlighting examples of dark patterns online since Harry Brignull, a UK user experience consultant, coined the word in 2010. But interest in online influence tools has increased over the previous year in the midst of a series of high-profile revelations about the handling of personal data by Silicon Valley businesses. An significant aspect of this debate is the concept of consent: what consumers agree to do and share online, and how far companies can go in guiding them to make choices.

• The prevalence of dark patterns on the web is unknown, but researchers at Princeton University started quantifying the phenomenon in a study published this week, focusing first on retail companies. The study is the first to examine a large number of sites systematically. The researchers created software that automatically scanned more than 10,000 sites and discovered that more than 1,200 of them used methods that the writers recognized as dark patterns, including false notifications from ThredUp.

The study coincides with debates among lawmakers on controlling technology businesses, including through a bill proposed in April by Senators Deb Fischer, Republican of Nebraska, and Mark Warner, Democrat of Virginia, that is to restrict the use of dark patterns by making some of the methods illegal and offering the Federal Trade Commission more power to police the practice.

“We’re focused on a issue that I believe everyone acknowledges,” Ms. Fischer said, adding that she became interested in the issue after becoming upset in her private experience with the methods.

The legislation faces uncertain opportunities, partly because of language defining dark patterns and the businesses that would be subject to the ambiguous new law, said Woodrow Hartzog, a professor of law and computer science at Northeastern University. Still, he added, discussing dark patterns is an significant first move for policymakers.

Many sites highlight the possibility that clients want to choose and play down options. The option to opt out of emails on one site studied by the Princeton researchers was in light gray type, making it look as if it could not be selected. Credit.

“As a policy issue, the important question is what separates a dark pattern from good old-fashioned advertising,” he said. “It’s a notoriously hard line to discover — what is permissible persuasion vs. wrongful manipulation.” Princeton’s research recognized web-wide dark pattern methods by automatically scanning the text and code of the pages.

For example, on ThredUp, the researchers saw the website create the April messages using code that arbitrarily selected combinations from a list of 100 names, 59 locations, and 82 items. The results were replicated by the New York Times. On one day this month, the code resulted to emails in which “Abigail from Albuquerque” appeared to purchase over two dozen products, including clothes in sizes 2, 4, 6 and 8. On other occasions, it produced messages depicting distinct individuals “just” purchasing the same second-hand item days or months apart.

When asked about the notices, a spokeswoman from ThredUp said in an emailed declaration that the firm used “real information” and that it included the fake names and places “to be vulnerable to privacy.” When questioned if the emails depicted actual latest purchases, the firm did not react.

According to Arunesh Mathur, a Princeton doctoral student and author of the paper, the number of sites found using dark patterns underestimates the overall prevalence of the techniques online. The researchers ‘ software focused on text and scanned only the pages of retail stores and not travel booking sites, social media services, or other areas where such tactics could be used. The research, he added, was also restricted to patterns used to affect buying conduct, not data sharing or other activities.

More than 160 retail sites used a “confirmshaming” strategy that requires customers to press a button that says something like “No thanks!” If they want to avoid signing up or buying something, I’d rather join the’ Pay Full Price for Things ‘ club.

More than two dozen sites used confusing messages to encourage users to sign up for emails and other services. For example, on a New Balance athletic apparel site, the first portion of a message proposed that a user could check a box to receive messages, but on closer reading, the reverse was true. “We’d love to send you emails with offers and fresh products,” he said, “but if you don’t want to receive these updates, please tick this box.” New Balance thinks the opt-out is “legally compatible and we think it is evident to customers,” said Damien Leigh, senior vice president of the company’s worldwide direct-to-consumer sales, in a declaration. But he added that the business “is always looking for ways to be as transparent as possible with customers and will assess the understanding of the study when it is released.” About 30 locations made it simple to sign up for services but especially difficult to cancel, requiring phone calls or other processes. The Times requires people to talk to a representative online or by phone to cancel subscriptions, but the researchers did not study it or other publishing sites.

Most researchers identified sites used messages that indicated that products were popular, that there were few items in stock, or that products would only be available for a limited time. Some were evidently false, while others were unclear.

There is disagreement as to whether messages about things like high demand constitute a dark pattern if they are true. But even those based on actual site activity are an attempt to play on the known weaknesses of consumers, said Arvind Narayanan, a professor of computer science at Princeton and author of the paper.

Org. pub. Nytime