Case study: Kraken Tech's generative AI tool for customer service

Case study: Kraken Tech's generative AI tool for customer service

1. Challenge:

Kraken is the technology platform behind Octopus Energy, contracted to serve 54 million customer accounts across the globe through licensees.

Going through a customer's account history and drafting an appropriate response can be time consuming. With diverse requests spanning from billing issues to technical support, responding effectively is resource-intensive and can lead to delays.

2. Solution:

Magic Ink is Kraken Tech's generative AI tool for customer service. Built on GPT-like models, customer service staff can use the tool to summarise all interactions with a customer, generate automated responses, and suggest actions, such as requesting a meter reading.

Magic Ink is fed continuous context around customer history and product specifications to enable customer service staff to efficiently deliver world-class service to customers. Magic Ink is integrated into the foundation of the Kraken platform, which is optimally placed to deploy AI:

1. Magic Ink has the maximum context available natively, with data from the entire customer lifecycle hosted on Kraken. Large language models (LLMs) make use of these large data pools.

1. Customer service staff are empowered to use Kraken to solve any customer issue. LLMs turn from toys to tools when they can actually take action to solve problems.

1. Kraken has been iterated and validated at scale - it's easy to make cool demos with AI, but the difference between demo and product in the AI sphere is huge. Our close relationship with Octopus Energy means that we understand the industry extremely well and the Kraken platform is tried and tested.

Magic Ink uses natural language processing to draft responses to diverse inquiries, as if written by the user. By reducing the burden on human agents, Magic Ink enables them to focus on more complex issues and answering phone calls, while providing swift and accurate assistance to customers.

3. Barriers:

• Hallucinations - To address this we have focused on a verification system which annotates verified facts with their source and highlights text which couldn't be verified. Team members are also trained to review everything written by Magic Ink with a fine-tooth comb. Around 50% of messages generated by Magic Ink are sent with no changes required from service agents.
  

• Training - We use AI as a tool to enhance individuals productivity, not replace individuals. As we've matured, we've learnt a lot about how best to onboard users and make sure they understand the ins-and-outs of the tool.
  

• Legal - Clients are understandably cautious about adopting it, so we test it extensively with our own teams and customers.

4. Impact:

Assisted by Magic Ink, our team has been able to deliver improved service and quicker customer response times. Around 35% of customer emails are currently written with the assistance of this tool, receiving higher customer satisfaction ratings (around 70%) than those without. This also drives employee satisfaction by automating routine tasks and freeing up teams' time to address more complex queries.

To date, Magic Ink has summarised 6,239,087 calls - the equivalent of 695,379 hours of talking time - and has generated a further 9,415,901 messages. In all cases, the system keeps human operators in the loop.

One user said: "It's been super easy to get to grips with and it's very clever how it adapts to your personal tone so quickly". Another shared: "I've been really impressed with Magic Ink's ability to create personable responses to our customers. It's a great tool for responding to short, snappy emails or a great starter for longer, more complex ones."

techUK - Putting AI into Action

Our Putting AI into Action campaign serves as a one stop shop for showcasing the opportunities and benefits of AI adoption across four strategic sectors - Health and Social Care, Cyber, Central Government and Transport.

techUK is coordinating a calendar of events, reports, and insights to demonstrate some of the most significant opportunities for AI adoption in 2024, as well as working with key stakeholders to identify and address current barriers to adoption.

Visit our AI Adoption Hub to learn more, or find our latest activity below.

Upcoming AI Adoption events

Latest news and insights

Get our tech and innovation insights straight to your inbox

Sign-up to get the latest updates and opportunities from our Technology and Innovation and AI programmes.

Contact the team

Learn more about our AI Adoption campaign:

Public link

Please use the above public link if you want to share this noodl on another website.