Vodafone Group

I worked alongside motion and visual designers, AI experts, and internal business SMEs to create a new customer communication service across responsive web and app, in addition to re-designing and standardising Vodafone Group's core sales and checkout journeys.

The UK's first telecommunication chatbot and Asynchronous Messaging tool, also known as 'TOBi', would prove to be highly successful through its smooth facilitation of customer service journeys and consequential reduction of call-centre reliance.


Vodafone Group


User Experience Designer




1 year


Customer service failures and outdated tech infrastructure.

Having been fined £4.6m by Ofcom in 2016 for failing to properly handle customer complaints, citing outdated technology and poor training, Chief Executive, Nick Jeffries, vowed to tackle the issue head-on.

The Telco giant would begin its mission to migrate critical customer support infrastructure by consolidating call centres into core hubs and, in relevance to this project, explore the use of artificial intelligence, machine learning, and biometric identification to improve customer support and sales experiences.

AI & Machine Learning.

Despite chatbots having been around in some capacity for fifty years, the recent rise in their usage can be largely attributed to one key development.

Artificial Intelligence has also made significant steps forward in the past few years. Associated technology such as machine learning, natural language processing and deep learning can dramatically improve the quality of understanding and decision-making.

IBM Watson for TOBi.

Watson Assistant was selected as the technology on which TOBi would be developed. As the industry-leading enterprise AI virtual assistant technology, it enables business users and developers to collaborate and build conversational AI solutions.


Define the UI framework and interaction model for conversational experiences at Vodafone.

A Familiar Customer Story.

Picture the scene. You’ve just bought a new phone on a Vodafone contract.

You get it home, tear away the packaging, slide in the sim and… it doesn’t work. After reading the instructions for the millionth time, you decide you’re not to blame, Vodafone is.

You decide to head online to see who can help and how this can be resolved as quickly as possible. After briefly reviewing customer FAQs without success, you become increasingly impatient. You experience a further 10-minute wait to speak to a customer services agent via Live Chat, who manages to troubleshoot the query with you and get you on the move again.

Not a good start, right?

This is just one of the user stories we designed as part of our vision to improve the customer experience across sales and service queries.  

Design Goal.

The goal was to define and design the optimal experience and interaction model for the conversational product suite, with specific consideration of its integration within the Group's tech ecosystem.

Product KPIs.

Project success would be measured by three performance indicators:

  • Containment: How many queries can TOBi can resolve without human intervention?
  • Handover: How many queries have to be handed over to a human customer services agent?
  • Failure: Query journey failure without successful containment, issue resolution or human handover.
Chatbot KPI Funnel
My Role.

My primary responsibilities included;

  • Design of a pattern library and interaction model for conversational UI that delivered the user experience via a familiar and easy-to-use paradigm.
  • Creating design guidelines to help local markets implement the technology.
  • Planning and facilitating usability testing sessions.
  • Providing implementation support to all international markets.


Understanding the main drivers for customer service queries.

Gathering data on customer intent.

As a first step, it was important to gather data that would help frame the problem, contextualise the landscape, set metrics, and establish a shared vision of what we’re trying to create.

What are customers doing when visiting help and support areas?

What assistance do they usually require and what questions do they have when navigating through core sales journeys?

The priority of customer journey design was driven by quantitative data on customer 'intent' such as call centre drivers and technical complexity.

Customer 'Query' Overview.
Help & Support Home: Click Map
Reviewing existing design paradigms.

Users will enter our service with a mental model that is influenced by their interactions with similar looking and feeling messaging experiences.

Four key features and themes were identified and would be core to any experience being delivered to our users.

User Input.

Text and Voice input are combined with Quick replies to facilitate task completion

Activity Status.

"TOBi is thinking" "Agent is typing"
An activity indication should be provided to the user to establish clarity on the system status.

Response Time.

The service should be clear and upfront on the expected response time. E.G "Typically replies instantly" for TOBi.


Be clear on who the customer is talking to. Is it TOBi or a Customer Services Agent? Messages from different actors in the conversation should be clearly distinguished.

Contextual Customer Interviews.

I conducted semi-structured interviews with existing customers to help us understand their thoughts and feelings when interacting with existing chatbot services in the market.

Key insights included the need to;

Avoid using chatbots in cases where the customer is likely to be sensitive or stressed, such as dealing with complaints.

Trained human agents are better able to express empathy to deal with the situation. We can use Natural Language Processing to identify these cases and divert them to human agents.

Clearly distinguish the ‘bot’ identity versus the ‘human’.

Several customers described situations where they believed they were talking to a "real person" and were shocked to find it was automated, resulting in a negative feeling of mistrust and caution about discussing sensitive information moving forward.


Component library, prototype creation and iterative user testing ahead of market launch.

Guidelines and Design Principles.

The team built upon the research learnings and created Group-wide guidelines on the core design principles and tone of voice, which was internally recognised as "The Three Laws of Chatbotics".

Do the task.

TOBi exists to help customers do things and so our design should effortlessly support that. All conversations should follow a familiar path:

Ideation Workshops.

Design workshops were used to generate ideas on the identity and interaction design for TOBi.

Crazy 8 Workshop
Component Library.

All designs would link back to a centralised Component Library that served as a framework for core conversational design patterns at Vodafone Group. This evolved alongside the journey design and user testing.

Component Examples
Interaction Model: TOBi & Asynchronous Messaging
TOBi: Visual Design
Prototyping and User Testing.

Designs were usability tested at low-fidelity and high-fidelity levels before releasing to international markets for local adaptation and development.

We were providing a blueprint that had room for localised nuances and differences in dev-framework.

I built early prototypes in Axure but this was the most commonly adopted tool for Vodafone and as the design became more stable, we coded the UI components to provide more realistic interactions.

Sample Results.

The sessions were predominantly task-based and users were both existing and new customers of a varied demographic.

The results were synthesised after every round of testing with the learnings directly influencing design tasks within the next sprint.

User Testing Result Sample: Billing Journey
User Testing Result Sample: Sales Journey


KPI results and TOBi today.

The activities TOBi would take over from its colleagues were the low-value, time-consuming, routine and transactional kind.

For BETA, I designed a UI framework that:

  • Helped customers quickly check their latest bills.
  • Provided an easy-access summary of usage.
  • Made it easier for customers to have their FAQ's, answered.
  • Assisted Q&A during new sales or upgrade journeys.
  • Allowed users to raise support tickets or complaints that could be prioritised and passed on to (Human) Customer Service Agents.
KPI Results.

We worked closely with different international markets to implement the Group's vision for the new service tool. In the UK, the results were remarkable.

  • Conversion rates with TOBi were between 60-100% higher when compared to standard, non-assisted, user journeys.
  • Transaction time was reduced by 50% when compared to non-assisted journeys with abandonment rates markedly lower.
  • TOBi received a usability score of >90/100 which was the highest a Vodafone service had ever received.
  • Net Promotor Score (NPS) >65 which improved on contact centre levels.
TOBi Today.

The chatbot has evolved into a service that is capable of offering immediate, relevant support to resolve more than 70% of customer queries and has doubled Vodafone website conversion rates.