The Role of AI in Social Media Analysis for Restaurants
Millions of reviews are posted online every single day. For restaurants, these reviews can make or break their reputations. Just a half-star increase in a restaurant’s online rating can have a profound effect on a restaurant’s revenue.
According to a study, more than 80% of a businesses’ valuable information can be found in the form of unstructured data such as reviews, Tweets and comments under social media posts. There is an infinite amount of valuable feedback available online which restaurants pay very close attention to. So, how do they do it? In my experience interacting with countless restaurant owners and managers, I have seen most of them attempting to analyse this feedback manually with a team of analyst. When you look at the volume and diversity of these reviews, you realise how time-consuming this is and how restricted the analytical capabilities are due to human limitation. In addition to this, it is critical for this information to be sorted and presented meaningfully and comprehensibly.
Platforms like Trip Advisor, Google and Yelp provide an overall score for each restaurant based on their reviews which is useful for consumers looking to determine the overall experience quickly without reading each and every review. But for restaurants, the key is in the finer details. The level of insights that can be abstracted from this gold-mine of information can help restaurants carry out evidence-based decision-making. Long waits for a table, bland food, poor service - how can restaurants extract these meaningful nuggets of information and filter out the noise.
Entity extraction and sentiment analysis are relatively new fields of research that implement artificial intelligent concepts to identify the subjects of a piece of text and identify the sentiment, i.e. positive, negative, or neutral. These solutions are fast, accurate and scalable, and particularly useful to restaurants as they can be used to generate insights for themselves as well as their competition in order to benchmark their performance.
By analysing every single comment written about a restaurant and its competitors within seconds, this technology can enable businesses to answer more sophisticated questions. How does my cappuccino compare with my competition? How does my vegan menu rank across the city? What are my competitors most talked about dishes? Which restaurants have increasing social media activity and why - over the last month? Furthermore, restaurants can search through retrospective data for specific topics. Time saved will allow restaurants to focus on other aspects of the business such as faster responses to negative feedback.
Using automated processes, I have even seen restaurants discover a plethora of insights that revealed a clear discrepancy in their perception and their customers’ perception of their brand. AI can help restaurants adapt their internal strategies with better speed and precision in order to improve their online ratings, develop more effective marketing campaigns, keep up with trends and improve overall customer satisfaction.
Ultimately, no matter how game-changing a technology is, if the insights are not presented in a simple, meaningful way and do not translate into business actions, the purpose is defeated. In this ever-growing, fast-paced, data-centric world, restaurants must understand the capability of such technology and leverage it to improve their business.
AI has transformed human work. The ability to filter through millions of reviews in seconds and extract comprehensive insights which can help with important business decisions is extraordinary. In an industry worth $900 billion it is a wonder why every restaurant hasn’t employed AI for its social media analytics. Of course, there are hundreds of solutions available on the market. Picking a hospitality specialist is vital; this gives restaurants contextual and actionable insights, relevant to the industry and the setup time could take as little as a few hours.
To those managers still wasting time online in the early hours of the morning with 30 tabs open, sifting through hundreds of reviews, trying to make sense of all the feedback, I’ve got one thing to say – get with the programme!