Cognitive Computing – Future of Retail?

So, what is really Cognitive Computing? In the simplest form, it is the simulation of human intelligence and thought processes using connected computer systems i.e, systems that can think, reason and learn with the goal to take an action. It uses various capabilities to achieve this – natural language processing, speech and tone analysis, image recognition, weather and location, to name a few.

Unlike yesteryears, today, it is highly likely that you will hear the word ‘cognitive’ in your professional conversations at least once a day (if not more) though the context in which people use ‘cognitive’ vary significantly. While researchers started discussing intelligent computer systems and cognitive computing in the 1950s, people started paying attention to the real world applicability of systems with cognitive capabilities in 2011, when IBM Watson, the super computer competed and won on the popular TV show Jeopardy.

So, what is really Cognitive Computing? In the simplest form, it is the simulation of human intelligence and thought processes using connected computer systems i.e, systems that can think, reason and learn with the goal to take an action. It uses various capabilities to achieve this – natural language processing, speech and tone analysis, image recognition, weather and location, to name a few.

Cognitive computing in retail – The Promise

Based on a recent survey by IBM, 83% of the retail executives believe cognitive will have critical impact to the future of their organization. While the promise of Cognitive Computing is to change the world, business leaders are a bit slow to adopt this into their day to day businesses. Even in a fast paced industry like retail, there are more proof of concepts than end to end business solutions offering cognitive capabilities. Let’s examine how a retailer could significantly up the game with Cognitive capabilities.

A recent Forester study of online users reveals some of their key pain points of shopping online:

  • 45% ofUS internet shoppers will abandon their purchase if they can't find a quick answer to their questions.
  • 50% of potential sales are lost because visitors can’t find what they are looking for or it is not personalized for them.
  • 29% of Online users prefer to use online customer service, during the purchase journey

This is one of the reasons why people still enjoy visiting a physical store nearby, interacting with the store associate to get personalized recommendations, pricing and promotions for completing the purchase journey.

How about combining the convenience of online shopping with the personalized attention and recommendations from the store sales staff? Sounds impossible or way too futuristic? Think again. This is exactly what cognitive capabilities in retail promises to offer - Understand the customer better in real time and inspire them with relevant product recommendations and content to enable enhanced business outcomes – be it better customer engagement, or better conversion. Many of the building blocks for this disruptive capability exists today. Key question is – are the retailers ready for adopting them?

Is cognitive a big deal?

Well, Yes and No. If your view of cognitive is limited to implementing Natural Language Processing (NLP), or a basic Chat Bot, it might give the consumer a differentiated experience but it might not really help retailers fulfill the promise of Cognitive. The journey needs to be made more rewarding by leveraging the capabilities of learning, intuition etc.

It is estimated that 80% of world’s data is unstructured. Typically, retail consumers add billions of tweets every day and similar number of Facebook posts on the products they reviewed, experience at a store and various other sentiments – largely in the form of text, images, and videos. In the Cognitive era, this chatter is one of the most important data set that can be used for proactively engaging them at an emotional level. Historically, most retailers focused on behavioral and demographic data of the customers to provide a differentiated customer experience. Unfortunately, this hasn’t achieved the level of personalization that the consumers were expecting. So, how can retailers leverage other data elements that are relevant to the consumers and derive insights to make the experience even more engaging? This is the opportunity that is knocking the doors of retail businesses. If we were to believe the forecast by leading players in the cognitive space, this market is estimated to be around $ 2 trillion globally in the next decadewhich is indeed a big deal.

So, how can retailers benefit from cognitive solutions?

Retailers have just started to scratch the surface by adding some basic Cognitive capabilities to their channels. Some common application areas that are becoming popular are – solution for information exchange using an automated human like agent (a.k.a ChatBot), providing ability to develop a conversation using natural languages, or a providing virtual way finder inside the physical store.

This is clearly just the tip of the iceberg. The real value of end to end cognitive solution will be realized once the retailers start looking at ‘Consumer touch points’ holistically and start using that data to make the experience hyper personalized. Nowadays, customers research well before buying products and services and they listen to their physical and virtual friends before narrowing in on options. Retailers need to listen in, be part of this journey and make intuitive recommendations. This would require some of the Cognitive platforms, data streams from relevant systems, learning algorithms, and more importantly, the business team’s ability to monitor the system’s learning and evolve it continually towards more accurate and effective outcomes.

So, where do we start? No doubt – it starts with understanding the concept of probabilistic decision making and the difference between probabilistic Vs deterministic systems. So far, business decisions were made based on deterministic systems that takes a certain input and based on the fixed logic or rule, an expected output is provided. This notion is changing with a probabilistic approach of cognitive systems where learning is at the center and system learning without being explicitly programmed for, is a new paradigm that IT teams and businesses need to get familiar with.

Who are the enablers and key players

Given the hype and opportunities, every major technology platform vendor and startup companies are jumping into the Artificial Intelligence / Cognitive bandwagon. IBM, Microsoft, Google, Amazon, just to name a few. IBM, with its much hyped Watson platform, has an obvious lead in this space, but it is only a matter of time before others catch up.

Even with all these platforms and point solutions, why are the retailers still far from the dream of hyper personalized experiences for their customers? And that is probably the biggest puzzle to be solved. Given that every retailer is unique in terms of the business models, positioning, target customers, IT systems etc and the problems that they are trying to solve are very specific, no ONE platform can solve it for them. And the answer lies in the retailer’s ability to stitch together an end to end solution that brings all of these together to solve it for consumers.

Recent News

Locations

46 Cities.
5 Continents.
1 Vision.