Retailers Keep a Close Eye on Today’s Changing Market with Sentiment Analysis

January 10, 2019 | Entity Extraction, Risk Management, Sentiment Analysis, Social Media Analysis

The holiday season is over, holiday decorations have been taken down, and everything is back to normal. Or is there “a normal” anymore?

According to preliminary retail sales figures, this was a very good holiday season for U.S. retail sales. Here are some highlights:

  • Record consumer spending. According to Mastercard, U.S. holiday retail sales grew at their fastest pace in six years to reach $850 billion. Total holiday retail sales from Nov 1-Dec 24 rose 5.1% from a year earlier.
  • The shift to online shopping continues. Online sales surged $19.1%. While department store sales slid 1.3% from 2017, department store online sales saw a 10.2% sales growth. Mobile devices (smartphone, tablets) accounted for about 58% of site visits and 39% of purchases.
  • The Saturday before Christmas, known as “Super Saturday” or “Panic Saturday,” outpaced Black Friday in store visits.

Despite a surge in consumer confidence and spending, the outlook for all retailers is not as bright as the sales figures from this holiday season would suggest. Brick and mortar stores continue to struggle in a changing market that has already pushed Sears Holdings into Chapter 11 bankruptcy. Competition for consumers is fierce. A record 65% of purchases between Thanksgiving and Cyber Monday were driven by promotions, which means less profit to be had. The customer journey has become more complex, with different shopping phases and patterns that crisscross the online and offline worlds. And last but not least, economic conditions are uncertain with ongoing volatility in the capital markets, rising interest rates, and stalemate in Washington D.C.

Retailers must keep a close eye on the market’s changing dynamics and consumers’ quickly evolving patterns and preferences to better position themselves, their messaging, and their marketing spend in their target channels.

Turning to the Online World for Insights

In the digital era, it’s the online world of social media, blogs, and forums that can provide the largest and most timely sample of ever changing consumer preferences and opinions. Here are various ways in which monitoring and analyzing online content is of critical value to retailers:

  • Market research and intelligence. Discovering what customers desire and value from retailers is a key insight in a fiercely competitive market, made even more competitive by ecommerce giants like Amazon and Ebay.
  • Brand promotion. Online content can provide valuable insight to assess how a marketing campaign like a new commercial or digital media ad is being received.
  • Reputation management. Knowing what is being said about a company is key for brand monitoring.  Consumers often turn to online reviews and comments when making purchasing decision. It’s important to maintain as good an online image as possible and in order to achieve that, companies must detect issues online promptly and publically address them.
  • Competitive analysis. Knowing how your company stacks up against its piers, and where it excels and misses the mark, is critical for a successful business strategy.
  • Customer care. Unhappy customers often turn to social media outlets like Twitter or online review sites to air their complaints. An appropriate and swift response can go a long way to appease an unhappy customer and prevent a complaint from going viral.

Sentiment Analysis: the Key to Better Understand Consumers

Sentiment Analysis, also known as opinion mining, is about detecting likes, dislikes, emotions, and opinions and is often applied to social media, reviews, blogs, forum posts, chats, and similar sources.  Sentiment Analysis is especially well suited for mining online content to extract insights into customer emotions and preferences. Here is how:

  1. Buzz. At the very basic level, Sentiment Analysis identifies the various mentions of a brand or concept or product and makes it possible to track those over time. Companies can monitor the number of mentions of their brand names, products, promotional hashtags, or slogans over time and be alerted if the number of mentions suddenly goes up. A high level of buzz could be good, but it could also be bad, as when an ad becomes unintentionally controversial or the object of widespread criticism, or there is a major piece of negative news about a company.  Similarly, comparing the buzz of your own brands with those of your competitors can give you insight into how effective you are in your overall target markets.
  2. Likes and Dislikes. Sentiment Analysis detects positive and negative sentiment around a brand or concept. At a very basic level, it identifies positive and negative language. At an advanced level, through Entity- and Aspect-based Sentiment Analysis, it pinpoints the specific aspects that those positive and negative sentiments are about. For instance, customers may love the latest store-wide sales, but be greatly disappointed by long lines, disorganized shelves, or poor customer service.
  3. Intent. Advanced Sentiment Analysis detects language suggesting intent and sentiment-based actions like intent to purchase or boycott as well as both positive and negative recommendations.
  4. Large Data Sample. Advanced Sentiment Analysis can scale to Big Data size, making it possible to process massive amounts of data in real time. Naturally, the larger the amount of data and the faster it can be processed, the more reliable, useful, and timely the insights and the corresponding responses will be.
  5. Analytics. Advanced Sentiment Analysis normalizes the extracted information thus enabling data aggregation and quantification to provide the overall view from a large collection of content. A dashboard presents multiple views of the sentiment information through various types of interactive graphs and charts. Sentiment data can be sliced and diced as desired, for instance by positive and negative aspects, and, if desired, the user can drill down to the source text for inspection and further analysis. Other useful charts show sentiment evolution over time through a sentiment timeline. Using location metadata and automated geocoding of mentioned place entities, sentiment can be plotted on a map to predict, for instance, where there will be higher demand of what products.

In summary, in today’s fast changing consumer environment, where there are so many market pressures and competition is fierce, retailers turn to online content for insights into consumer and market trends. Sentiment Analysis is the AI-based technology for the modern age. It helps retailers perform more cost-effective and accurate market research, brand reputation monitoring, customer care, and competitive analysis.