What do Google search queries reveal about brand attitudes?
It has been widely thought that brand search volume — the counts of queries that a search engine receives from users that include specific brand names, like "iPhone" — can predict sales for that brand. Google makes this information available to marketers for free through Google Trends, an application that allows anyone to see how frequently popular brand names are searched, and it has become a valuable tool for marketers. But new research from Drexel University, Brigham Young University and Google shows there is a lot more to consider when looking at the large number of queries that include brand names.
In order to investigate the relationship between brand attitudes and search engine queries, the researchers — Jeffrey P. Dotson, associate professor of marketing and global supply chain at Brigham Young University; Elea McDonnell Feit, assistant professor of marketing at Drexel University; and Ruixue Rachel Fan, Jeffrey D. Oldham and Yi-Hsin Yeh of Google — studied over 1,500 Google users who opted in to have their searches related to smartphones and vehicles tracked over a period of eight weeks and then linked their responses to a traditional brand attitude survey.
"Using a panel of users who gave us specific permission to track their brand searches gave us a unique perspective on how brand attitudes relate directly to what individual users search for," said Feit. "It was a lot of work to recruit that panel and secure their permission, but the end result is some findings that we think are vitally important to marketers."
They found that users who are actively shopping in a category are more likely to search for any brand. However, as users move from being aware of a brand to intending to purchase a brand, they are increasingly more likely to search for that brand — with the greatest gains as customers go from recognition to familiarity and from familiarity to consideration. For example, someone who is just beginning to consider purchasing a car, is more likely to start with a general search that does not include any brands. As the consumer narrows their decision before a purchase, he/she will start searching for a specific brand.
They also found that users who own and use a particular car or smartphone are much more likely to search for that brand, even when they are not actively shopping. This suggests that a substantial volume of brand search in these categories is not related to shopping at all — something marketers should keep in mind if they see a sudden spike in search traffic for their brand.
Positive brand attitude and consumer searches
Specifically, in the smartphone category, the estimates from the research indicated that all five of the brand attitudes–recognition, recall, familiarity, purchase considerations and purchase intent–were positively associated with brand search. The data confirmed that customers who hold positive attitudes toward a brand are more likely to search for that brand. For example, in the smartphone category the odds of searching for a brand is seven times higher for a user who holds all five positive brand attitudes, versus a user who holds no positive brand attitudes.
Similarly, for the automotive category, the odds of searching for a brand for a user who holds all five positive brand attitudes is five times higher than for a user who doesn't hold any positive attitudes. For both categories, they found the biggest increases in search propensity to search from customers who are familiar and would consider purchasing the brand.
Customers who only recognize a smartphone brand and hold no other positive attitudes, were only 1.22 times more likely to search for a brand than those who don't recognize the brand.
Behavior of customers in the market
The researchers found that customers who are actively shopping are more likely to search for any brand in a category.
"Customers who indicated that they were in-the-market during the observation period were significantly more likely to search for any brand," said Dotson. "Similarly, customers who indicated that they 'always pay attention to the category so that they know when to buy' are more likely to search for all brands in the category. This suggests that a substantial portion of the brand search queries that are submitted to Google are associated with users who are shopping for the product."
Customers who own a brand
The researchers found that owning a particular smartphone or vehicle brand is a very strong predictor of brand search, with the odds of searching being 2.5 times greater for brand owners versus non-owners in the smartphone category and 3.5 times greater in the vehicle category. The large increase in brand search among owners (regardless of whether that user is actively shopping), could be partially due to owners searching for information about how to use the product, according to the researchers.
Marketers who are interpreting total brand search volume (e.g. Google Trends data), should expect that brand search will be higher for brands with more owners, irrespective of consumers' attitudes toward the brand.
"Somewhat surprisingly, we did not find that owners who are having problems with the brand are significantly more likely to search for the brand than other owners, suggesting that information search is not simply associated with problems or product recalls, but that there is a steady volume of search produced by brand owners, where consumers are likely to have questions about usage and maintenance of the product they use daily," the researchers wrote.
Product category effects
It's also important to note that some categories of products will always draw more searches before a purchase because of the nature of the purchase. Research prior to purchase is likely to be greater for furniture, appliances, financial services, cars and smartphones. For these categories, an increase in brand attitude could lead to increases in search queries for the brand. For other product categories, searches prior to purchase may not apply but that doesn't mean that purchases are minimal. Coca-Cola, for example, may not have a high search volume in queries, but has a high volume of sales.
"There is still a good deal of work to do to figure out exactly how marketers should interpret brand search volume for specific brands in specific categories, but this research confirms that marketers should start using brand search volume data to monitor their brand's health," said Dotson.
Feit added, "Marketers have long used brand attitude surveys to monitor brand health but surveys are expensive and fewer and fewer people want to answer them. This research adds to the mounting evidence that we can track brands more passively using new data sources like brand search."
The paper, "Brand Attitudes and Search Engine Queries," is forthcoming in the Journal of Interactive Marketing.