According to research from USC Marshall published in 2022, consumer-posted photos are a leading indicator of restaurant success.
While the fate of a restaurant might not hinge on a single consumer posting or viewing photos, consumer-posted photos collectively provide valuable information beyond reviews, restaurant characteristics, competitive landscape and macroconditions, such as year and zip code.
“It takes very little effort to upload photos on social media. As such, photos may serve as a proxy of restaurant popularity above and beyond reviews,” said Lan Luo, associate professor of marketing at USC Marshall. “We also conjectured that photos may help viewers better visualize how much they might enjoy the food items and/or the dining experience.”
Luo and Mengxia Zhang, who earned her doctorate at Marshall in 2021 and is now assistant professor of marketing at the Ivey Business School at Western University, published their findings in “Can Consumer-Posted Photos Serve as a Leading Indicator of Restaurant Survival? Evidence from Yelp” in the academic journal Management Science and filled a major gap in the research on restaurant viability.
“To the best of our knowledge, this is the first empirical study that explores whether consumer photos can serve as a leading indicator of long-term business prosperity,” Luo said. “Our research is also among the most comprehensive empirical studies on restaurant survival to date.”
According to the National Restaurant Association, the U.S. restaurant industry generated $799 billion in sales in 2021, down $65 billion from 2019’s pre-pandemic level. The industry employs 14.5 million people, down 1 million from pre-pandemic levels. Yet there is scant large-scale empirical research on (non-COVID-related) restaurant survival.
The COVID-19 pandemic has made this research even more relevant to the food service industry. “Consumer-posted photos could be more important for restaurants in a post-pandemic world, given that many consumers order food online now and that photos can be really helpful for consumers to figure out whether they may like a food item,” Luo and Zhang said.
The groundbreaking study employed machine learning techniques to extract features from 755,758 photos and 1,121,069 reviews posted on Yelp for 17,719 U.S. restaurants. The analytics tools also factored in restaurant characteristics (cuisine, price) and competitive landscape as well as entry and exit (if applicable) time.
Read the full methodology here.
The study found that photos that featured food had the most positive associations, followed by interior and exterior shots. The proportion of photos with helpful votes also positively relates to restaurant survival.
The potential applications of this research are far-ranging. “Multiple stakeholders can readily apply our proposed framework to improve their decision-making processes,” the authors wrote. “Findings from our research can be beneficial to restaurant owners, business investors, landlords, online platforms and research/trade associations.”
