In today’s customer-centric environment, there is no shortage of outlets for consumers to express their satisfaction with a company, employee, product or service. At some point, I am sure you have gone to an online review site such as Amazon, Rotten Tomatoes, TripAdvisor or Yelp to express your level of satisfaction or dissatisfaction with a specific product, service or organization.
 
Such feedback is vitally important to business firms, as customer reviews are shown to be more trustworthy than descriptions that come direct from manufacturers. For instance, 73 percent of people trust online reviews and 63 percent of people actively seek out online reviews when making purchase decisions. Moreover, social media sites like Facebook, Instagram and Twitter allow us to share our experiences with a company, brand or product with family and friends. This is shown to exert a great deal of influence on our subsequent purchase decisions. How many times have you have used your social media accounts to get information and recommendations for service providers in the greater Wilmington area?
 
Yet most customer feedback data is free-form text. This generally implies that the data is non-numeric (qualitative) in nature, which makes it difficult to structure, organize and transform into numeric (quantitative) data. While many online customer feedback sites provide graphic rating scales (stars, tomatoes or smiley faces, for example) to convey an average quantitative rating (such as 3 ½ stars), they only provide a “snapshot” view of the data. In other words, it’s simply an average rating for a product or service at a single point in time.
 
How much information value does that really provide? If I were visiting the greater Wilmington area, it seems my decision to patronize a restaurant would be more informed if based on how that average rating has changed over the past year. In addition, many restaurants in Wilmington are staffed by UNCW students. However, many of these same students leave Wilmington at the end of the school year, requiring local restaurant owners to hire new staff during the busy summer months. Would this seasonality effect influence the average rating? For instance, might a family on their summer vacation in Wilmington get less-than-average service when the restaurant lacks experienced staff? These are the type of questions consumers should be asking and business owners need to be addressing.
 
Fortunately, there are many free and commercially available software tools that can be used to quantitatively analyze and interpret customer feedback data. Here are a few:

• Semantria: https://semantria.com/
• Repustate: https://www.repustate.com/
• Text2Data: http://text2data.org/
• QI Macros: http://www.qimacros.com/
• Twitter Sentiment: http://twittersentiment.appspot.com/
• Meltwater: http://meltwater.com/
• Trackur: http://trackur.com/