How we design the tools that power our award-winning customer support experience

Read the article

At Monzo, our Operations department (or ‘Ops’ for short) is responsible for our customer support team and the in-app help experience. This means the scope of an Ops designer includes our customer-facing products, like our apps and website, as well as the internal tools, systems and processes that our support team use every day. And with that, comes some significant challenges.

In this blog post we’ll dig into how we approach and tackle these challenges to design a great help experience. We’ll be covering five areas, including some examples from recent projects we’ve worked on:

  1. Designing for two audiences

  2. Handling stressful situations

  3. Thinking in systems

  4. Scaling our solutions

  5. Learning from data and research

Designing for two audiences

Having a deep understanding of our users is an important first step in designing a great user experience. In Ops, we have two very different audiences to understand; our customers and our customer support team (or ‘COps’, as we call them). We like to think of our COps as customers who use our product too, but in their case the product is our internal tools and processes. Because our projects often span both our public-facing and internal products, it’s important we think about both audiences in terms of their motivations, capabilities, and experience.

Our in-app fraud reporting feature is a good example of this. To help a customer report fraud, we need to understand their capabilities in order to give the appropriate level of guidance. For example, explaining card blocking to customers who might have never experienced it before.

Once a customer has reported fraud, a task is created for a COp to review the case and decide on a course of action. To help the COp make their decision, we need to understand their experience level in order to show the appropriate information. For example, summarising and highlighting the key facts of the case to guide less experienced COps.

A screenshot of the internal tooling we use to handle Fraud disputes, next to the in-app experience a customer sees when disputing a transaction

Handling stressful situations

The mental and emotional state of a customer who needs support tends to be more delicate and warrants special attention. For example, a customer who needs help is likely to be confused or frustrated, and a customer who wants to complain might be angry or upset. So we need to be very thoughtful with how we handle them, giving more reassurance and building trust wherever we can.

For our fraud reporting feature, the approach we took was to map out the customer’s emotional journey. By considering how their feelings and stress levels might change throughout the experience, we were able to pinpoint key moments where we should increase the feeling of trust or add more guidance.

These findings directly informed the design of the experience. For example, reducing stress by reassuring the customer that their account is safe, and increasing clarity by setting expectations about the process and what happens next.

The emotional journey a customer might go on, from a place of most stress when initially worried about an issue, to less stressed once they've engaged with our support

Thinking in systems

As designers, it's often tempting to dive straight into pixels when a problem seems simple on the surface. But before jumping to solutions, it’s important to understand the scope of the problem and how any planned changes might impact the surrounding environment.

In Ops, the tools we build tend to be underpinned by complicated systems with many interconnected processes. These processes often involve the handling of lots of data. So understanding how the entire system works and how data moves around the system is often key to understanding the whole problem space. This allows us to spot weaknesses and opportunities for improvement.

In a recent project to improve how we help customers who’ve been a victim of impersonation fraud, it was important to not only map out the customer’s journey in the app, but also the steps that our COps would need to go through to help the customer, including 3rd party involvement such as the police. By mapping this out on a systems level we set a foundation of understanding for the whole product team. This enabled us to spot opportunities for improvement within the existing processes before diving into design details.

An example of how we plan out a customer journey

Scaling our solutions

As a small design team working on a complex product, the solutions we create in Ops need to be scalable and adaptable to work in many different situations.

Designing UI components in a modular way gives greater flexibility. This also makes it easier for engineers to build a reusable system, rather than creating bespoke solutions every time. The end result is that other teams can use the same patterns, saving time and money when tackling similar problems and giving a more consistent user experience.

As part of our work to create an in-app flow for disputing transactions, we needed a structured way for customers to answer questions about their case and upload evidence. We didn't want to build a solution that would only work for disputes, so we focused on creating a modular form pattern. This has enabled other teams across Monzo to create their own self-serve experiences with a consistent UI, made from our simple building blocks.

Three screenshots of disputing a transaction in Monzo, starting with what happened, how you've contacted the merchant, and finally entering details of what you ordered.

Learning from data and research

Data and research are both super important in helping to understand a problem and inform our decisions. In Ops, we lean heavily on quantitative data, and focus in with qualitative research and testing whenever possible. There's 2 main reasons for this.

Firstly, we have a lot of data, and lots of data scientists! Lots of customers and COps use our systems every day, so it's relatively easy for us to gather large amounts of data and react quickly to the signals we see.

Secondly, it can be difficult to get accurate qualitative insights about customer behaviour in sensitive or urgent situations. In a user testing environment it's impossible to recreate the feeling of being stressed or upset, which will affect how a customer interacts with our product. It also wouldn’t be realistic or ethical to ask a customer to wait when they have a problem so we can observe their behaviour. When we need qualitative insights, we carry out foundational research to understand customer attitudes, or retrospective interviews and surveys to gather sentiments.

For anything to do with our internal product, we're very fortunate to have direct access to our COps. This makes it easy for us to understand how they work and learn about the problems they face. Shadowing is a great method for doing this, and it's amazing how much you can learn in a short space of time just by watching someone work.

The methods and techniques outlined here have helped us balance an empathetic customer and COp experience, with the need to design at scale within the bounds of a complicated system. We’re not done yet though! Our design process is evolving, as we continue to learn about our users and make improvements based on our past successes and failures. If you’re designing a customer support experience, or working in a team that faces a similar challenge, the methods outlined here might help you approach your problem space in new ways.

Help us design Monzo

Are you a designer or researcher looking for a new challenge? Then you’re in luck! We’re currently hiring, so check out our careers page for more information about our open roles and to apply.