How Ratings and Reviews Impact Search

Reviews have always been important to the customer experience. Customers are more likely to trust businesses that have positive reviews, and around 89% say they read reviews before making a purchase.* However, a benefit that typically isn't associated with reviews is their impact on search. Reviews impact search in a number of different ways. These experiences vary depending on whether someone is doing a branded search query. Branded searches are searches for a specific brand (e.g. Starbucks near me) while unbranded searches don't feature brand-specific terms (e.g. Coffee near me). Let's start with what marketers care the most about: showing up for unbranded searches. Google's main criteria for how they rank businesses in search is based on distance, relevance, and prominence.** A big part of prominence consists of your online reputation. By definition: average rating, the number of reviews, and the recency of your reviews all contribute to make up your online reputation. On top of that, Google heavily leans on average rating when determining what the "best" business is. Take this common search query: "Best coffee shop near me." How do you think Google knows what the best coffee shop is? Google probably isn't going to every coffee shop and trying the coffee (sign us up though if a job like that ever opens up!). Instead, Google uses average ratings to deliver what they believe are the "best" results. Google often curates the results to only show businesses that are rated 4 stars and above when users include the word "best" in their search.*** That's right, in this case if you're a coffee shop rated 3.9 stars or lower, you don't even have the opportunity to show up. In addition to average rating, Google also utilizes the content of reviews to help justify why a business was ranked. Using the same example as above: "best coffee shop near me" – Google will crawl the review content for each business and highlight the keywords in bold that match the search query. aThe top three coffee shops in this example all have snippets from a review (known as review justifications) that contain the keywords "best coffee." Now let's look at how reviews impact branded search experiences. Take this search for example: "Joe Smith Canuck Insurance" This is the search engine results page on Google, with Joe Smith's business profile on display to the right. Google will crawl the web for all types of data, but will sometimes pull in information from other places to enrich their own search experiences. An example of that is on display here with Google prominently displaying ratings and reviews from other sites directly on Joe Smith's Google Business Profile. One of those review sources happens to be first-party reviews (reviews that Joe Smith generated himself and then published on his website page). A benefit of generating first-party reviews with proper schema markup: Google will sometimes reward businesses by showcasing the star rating at the top of the search engine results page (as seen in the top left corner). Fully understanding how much ratings and reviews have an impact on search can be eye opening for most businesses. The next challenge is understanding how to implement a reputation management strategy. Here are some strategies and best practices that our most successful customers have implemented: Monitor reviews. Knowledge is power, and the first building block for any reputation management strategy is to have visibility into your brand's ratings and trends over time. This can help you keep a general pulse on customer sentiment towards your brand's service. But looking at ratings can sometimes be misleading; here at Yext we recommend using more advanced tools to understand the specific reasons why ratings are trending positively or negatively. This is where sentiment analysis comes into play. Yext's Sentiment Analysis tool uses Natural Language Processing to identify important keywords in your customer reviews and assign a positive or negative sentiment score to each keyword. This functionality allows you to analyze your reviews based upon the actual content of each review and understand how your customers feel about different aspects of your brand. These insights provide a better understanding of your customers, helping you to deliver an excellent customer experience! The problem with most review management vendors is they can only capture high-level sentiments; they can't do keyword-level tracking and those that do, can only group keywords into one of three buckets: positive, negative, or neutral. This level of analysis, however, doesn't account for extreme sentiments. At Yext, sentiment scores can range between -100 and +100, with lower scores corresponding to negative sentiment, and vice versa. Let's look at two keyword examples that both have a negative sentiment: "waits" and "floors." We can see "waits" has a sentiment score of -12 because people are moderately displeased with the long wait times, but "floors" has a significantly worse sentiment score of -80 because people are fiercely unhappy with the dirty floors. Using the sentiment analysis provided by most review vendors (as described above), both keywords would fall into the same "negative" bucket.