Published in Platform content on June 2017. 4 minute read

Using Machine Learning to improve customer service

  • Home
  • Using Machine Learning to improve customer service

You might think Machine Learning is only for organisations with large amounts of complex data who are looking for deep, strategic insights – but that's not the whole picture. Machine Learning also has a role to play in helping organisations like yours deliver better customer service right now, sometimes using relatively small amounts of data.

If you're looking for inspiration when it comes to using the cloud to enhance customer experience, download our free how-to guide.



Take online grocer Ocado. Its contact centre is using Google's natural language API and a Machine Learning model developed in TensorFlow to automatically tag incoming emails based on their content. Ocado can then determine from the tags how urgent each email is and prioritise it in the queue, rather than handling everything on a “first in, first out” basis. This is helping it to deal with spikes in email traffic, which can double or triple when deliveries are affected by external factors such as bad weather or traffic jams, and allowing it to respond more quickly to customers who have an urgent request or are unhappy. Urgent emails are now dealt with four times as quickly as less urgent ones, and three times as quickly compared with the previous approach. Ocado has also been able to redeploy 4 FTE staff because it no longer has to review every email manually before processing it.

Another company looking to use ML to improve customer interactions is global bank ING, which has developed a Facebook Messenger bot using the Google API.AI conversational user interface and Google’s natural language API. Called Marie, the bot allows the bank's customers to get help as if they were chatting naturally with a human customer service agent, without having to wait for an agent to become available.

Meanwhile, online marketplace eBay is using ML to improve the customer experience for both shoppers and sellers. For example, the eBay Shopbot provides a personalised shopping assistant that can learn shoppers' preferences, expressed using natural language, and recommend items from its catalogue of around one billion live listings. For sellers, eBay has created an application that allows them find out the value of items they might want to sell, simply by talking with their Google Home device and answering a series of questions about the item's make, model, condition and so on.

These are just a few examples of how machine learning is changing the way companies interact with customers. If you’d like to explore how machine learning could help your organisation improve its customer service, please contact us to talk to one of our Google Cloud Platform specialists.

Download 6 ways to enhance customer  experience with Google Cloud

Free resources

Please download any of our resources to help with your research and project specifications