Cognitive

Tesco Hack 2017: Offering Personalised Promotions to Customers

Share this post:

 

 

Team IBM, formed left to right of Kiran Smith, Peter Ford, Hugo Pinto, Zack Pollard and Dan Cunnington attended the annual Tesco Hackathon at Tesco Labs in Welywn Garden City. The team started by using IBM Design Thinking to focus on two particular pain points for Tesco and their customers. Firstly, promotions to customers are currently tailored to the customers purchase history, but they aren’t always relevant given the current context. As a customer this can be frustrating, if you are offered a product which you are not likely to purchase. Secondly, it can be difficult for a large retailer to manage stock and waste in realtime leading to many items being reduced in price or thrown away to rubbish.

 

In order to solve these pain points, Team IBM decided to predict personalised promotions for customers given their current situation or context and offer these promotions to customers in real time using a Tesco mobile app. Not only does this provide relevant promotions to the customer, it also allows Tesco to manage stock and waste more efficiently.

 

To showcase our solution, we built three components.

 

  • An Android mobile app display the appropriate promotions to customers and allow them to scan and apply promotions as they shop.

 

  • An Angular.js dashboard to analyse spending behaviour by various customer demographics on a store by store level.

 

  • A TensorFlow Neural Network to predict relevant promotions.

 

 


Predicting relevant promotions

In order to predict relevant promotions, we trained a TensorFlow Neural Network using the following features:

{ Time, Nearest Store, Weather, Social Sentiment from IBM Watson Personality Insights, Set of products purchased }

Given the current time, the customers nearest store, the current weather conditions and a social media sentiment score, the neural network outputs a set of promotions that the customer would likely be interested in.

Thanks to Tesco Labs for a great event, if you would like to find out more about the solution please email Dan Cunnington – dancunnington@uk.ibm.com

More Cognitive stories

Protected: Raise the Game: Cognitive Analysis in Sports

There is no excerpt because this is a protected post.

Continue reading

Rainbow Octopus – Animate a 3d character with ARKit & IBM Watson

Rainbow Octopus is an open-source project created to show how a developer could use IBM’s Watson services in ARKit and Unity to control and manipulate a 3d animated character. Specifically, via the Watson Unity SDK, we capture and send speech to the IBM Speech To Text, and interpret the results using Tone Analyser and Watson […]

Continue reading

Finding Hidden Cell Phones in Prisons

Myself and some colleagues from IBM’s Emerging Technology attempted a two-day project just before Christmas this year, because we thought we might have the tech to tackle a big problem: cell phones in prisons, we know prisoners have them, but they are rarely ever tracked down. This is something the UK government is currently planning to spend […]

Continue reading