Whether you’re crashing at a stranger’s home or getting a ride in their car, participating in the sharing economy involves a substantial amount of uncertainty and an eventual leap of faith on the part of the consumer. When booking an Airbnb, for example, you can learn only so much before you decide to take a chance and stay there, sometimes sharing the space with a host you’ve never met before.
You’re not flying totally blind, however. Airbnb listings include some knowable facts about where you’ll be spending the night. You can see the property’s price, location, and amenities, and you can check out other customers’ reviews. These details, known to researchers of consumer behavior as informational cues, make up a good portion of the investigation a consumer does before booking their stay. The rest involves a typically less quantifiable process of sussing out the overall vibe of the place. Does the host look warm and inviting? Will this be a pleasant place to spend the night? This involves relying on noninformational cues, which provide no explicit information but instead appeal to emotions, humor, or aesthetics. For her recent paper published in the Journal of Consumer Research, Elizabeth Friedman, assistant professor of business in the Marketing Division at Columbia Business School, examined the effect of one specific noninformational cue on Airbnb listings: the host’s smile — or lack thereof.
In customer service, smiles are a known boon, at least to in-person interactions. Prior research has shown smiling increases perceptions of warmth and improves the mood of both people in the interaction, leading to greater consumer satisfaction, higher employee evaluations, and repeat customers. The new study explores the effects of smiling in an e-commerce setting, where face-to-face interaction is impossible.
How Smiles Impact Consumer Choices
To quantify the effect of an Airbnb host’s smile on property demand, Friedman and her co-researchers used large-scale observational data gleaned from more than 9,000 hosts across eight months in seven major cities, including New York, Los Angeles, Austin, and Seattle. They relied on machine learning to automatically identify whether a host was smiling in their profile picture and then estimated an econometric model to track demand for the property, measuring demand by how frequently the property was booked. To control for other variables among the properties — like the price, the host’s level of experience, and the neighborhood — the researchers applied inverse probability of treatment weighting (IPTW) to create a matched sample where the smiling condition of the host was separated from other variables.
They found that when a host smiles in their profile picture, it increases demand for their property by about 3.5 percent on average. Multiple variables moderate this effect: The impact of a smiling profile picture is greater for listings in higher-crime areas, for shared accommodations (where guests are more likely to interact with the host), and for hosts who are inexperienced or male (prior research has shown men are perceived as less warm than women).
The researchers theorized that each of these variables creates greater degrees of uncertainty regarding the quality of the accommodations and the nature of interacting with the host and that the smiling profile picture helps allay some of that uncertainty. “Smiling creates a positive halo effect for the host, which increases perceptions of the host’s warmth and competence, thereby mitigating uncertainty and increasing demand,” they posited in their paper.
The researchers also conducted four controlled experiments to further validate their findings from the observational study, surveying hundreds of participants on their uncertainty about booking Airbnbs and on how they responded to listings with photos of smiling versus unsmiling hosts. The experiments confirmed male hosts caused greater uncertainty for users surrounding their anticipated interactions with the host. They also demonstrated that smiling hosts appeared warmer and more trustworthy and competent to users and increased the likelihood of booking, particularly when the hosts were men. “This is consistent with our theoretical framework suggesting that there is greater uncertainty interacting with male hosts at baseline, and a resulting greater increase from smiling” compared to when the hosts were women, the researchers said.
Beyond Airbnb
Notably, in the fourth experiment, Friedman and her co-authors extended their research beyond the shared economy, focusing on hotel bookings. Nearly 800 participants rated their likelihood of booking hotels where a picture of the manager was shown smiling or unsmiling. The managers were presented as working for either family-owned boutique hotels or a Hilton. Smiling increased participants’ likelihood of booking at both types of hotel, but the effect was greater for the smaller, family-owned hotels, where researchers say uncertainty about the experience could be higher than when booking with a major hotel chain.
The study’s results have direct applications across the hospitality industry, quantifying how noninformational cues like smiles can impact consumer behavior, even online. The findings show when a smile is most effective, “helping practitioners optimize their online visual presentation and increase customer engagement,” according to the paper.
Furthermore, the study opens up new possibilities for future research, suggesting that noninformational cues could be relevant in other domains where people first connect online before meeting in person. The same theory could be tested among lawyers or physicians, who often include their photograph on their website as an initial introduction to a prospective client. For that matter, the researchers say, people using dating apps similarly begin their interactions online, where their prospective date scours their profile and photographs for noninformational cues that they’d be a pleasant person to meet for dinner. “These photos often serve as an entry point in a communication channel, so a smile may be a way to reduce uncertainty regarding the anticipated service or interaction,” the study states.
Adapted from “Serving with a Smile on Airbnb: Analyzing the Economic Returns and Behavioral Underpinnings of the Host’s Smile” by Elizabeth Friedman, Columbia Business School; Shunyuan Zhang, Harvard Business School; Kannan Srinivasan, Carnegie Mellon University; Ravi Dhar, Yale University; and Xupin Zhang, University of Rochester
Read more about the research here.
