May 30, 2018
Content Strategy in Financial Corporations — An Interview with Lauren Lucchese of Capital One
Lauren Lucchese is the Head of AI Content at Capital One.
Lauren’s team designs conversations for the Capital One skill on Alexa and Eno, the first natural language SMS chatbot by a U.S. Bank. She is also a mentor at Designation, a UX bootcamp, and co-organizer of Chicago’s UX Strategy Meetup, which explores how UX creates value for organizations.
[VG]: Lauren, I am thrilled to have you here in this ‘content strategy’ series. First thing first, I am in love with how I can read your role as “Head of AI Content”. Tell me your primary work in this role, your position in the team, and also some background for how the product team came up with this role.
[LL]: Thank you so much for including me in the series, Vinish!
As Head of AI Content, I built and now lead a team of content designers that create the conversations our customers have with Eno, Capital One’s intelligent assistant, and with our skill on Alexa. My team is also responsible for evolving the design of Eno’s character, which is a critical aspect of our design work (more on that later.)
We’re part of a larger Integrated Experiences team at Capital One, led by Steph Hay. Together, we’re working to bring humanity and clarity to every experience we design.
[VG]: Well this is so exciting. Please continue for more…
[LL]: I actually joined Capital One three years ago as a UX Content Strategist. A trained UX designer but a writer at heart, my natural inclination has always been to focus on the words in any of the experiences I was part of designing. As a UX Content Strategist, then, my primary goal was to demonstrate the power of words to deliver the right message, to the right person, at the right time, in order to create meaningful, tailored experiences that felt like personalized conversations.
That goal remains the same for the work I am leading in my current position. Now, we’re just faced with an added, fascinating design challenge:
What happens when the interface goes away?
The content is now the experience, and the words are your interface. The words you choose to use are all you have to build a meaningful connection and develop trust, so the pressure is on to choose wisely. Additionally, in designing for Conversational UIs that utilize AI technology, the more we can find the natural language that customers use to talk about money and design for that, the more we’ll have enabled a new kind of interaction that feels more true to real life for our customers.
Our design team has always operated with an overarching goal of meeting our customers where they are, and these new Conversational UI platforms create entirely new and exciting ways of accomplishing just that. In designing for our skill on Alexa and now Eno, our team saw an opportunity to take the conversation design principles we’d already been applying to our work and capitalize on an opportunity to create another meaningful communication touchpoint with our customers.
Now, our customers can get the information they need about their accounts by chatting with Alexa in their kitchen while they’re doing the dishes, or texting with Eno while they’re waiting in line at the grocery store. It’s pretty exciting stuff.
[VG]: ENO is so interesting — here is a quick look at the promise.
[VG]: Impressive indeed. The Capital One design culture is quite “research intensive”. Ryan Page, Head of Design, Card Partnerships, was quoted in this InVision Design Genome Report, as — “Many of the facilities we’ve built have actual labs in them. A lot of the research is self-conducted, in house, observing users.” Since you are in AI Content — tell me how this research culture guides the content processes in Capital One.
[LL]: Research provides us with a deeper level of understanding when it comes to our customers and the problems we want to solve, for them and for the company. Without it, we’re making decisions based on our own assumptions and best guesses, which is obviously never ideal.
When it came to creating Eno, our Conversational AI team conducted extensive user research and paid close attention to the feedback from our pilot users, and the lessons were many; we learned early on the value of doing a few things really well, versus many things so-so, and failing gracefully when Eno can’t understand what the customer is asking. As I mentioned before, one of our main design challenges here is to uncover the most natural language for that experience (i.e., the actual words customers use to describe what they want to do.). There are multiple ways we can gather this kind of data, but one of the most important remains engaging our customers in emotional conversation early on in the design process.
One of the major promises of Conversational UIs is the AI technology that powers them. Namely, the system’s ability to learn from every interaction we have with a customer. But data on its own does nothing by way of connection.
So much is baked into how and when that data is presented to the person. It’s stellar design and UX COMBINED with data and algorithms that make a product feel smart.
Part of this intelligence, then, is predicated on programming algorithms to deliver experiences that get to the heart of what people want and need to know, when they need to know it, which necessitates knowing the people you’re talking to and understanding how they want to be communicated with.
To successfully design in the AI space, understanding our customer’s perspectives and earning and maintaining their trust is critical. There is so much that is unknown and misunderstood about AI technology — what it is, how it works, and what it is capable of.
Much of our ability to design meaningful interactions for our customers in these channels requires understanding peoples’ boundaries when it comes to these types of experiences, which is something we explore through different forms of user research.
There can be a fine line between an interaction where it feels like a stranger knows a little too much about you, and an interaction where you connect in a meaningful way because it feels like what you’re interacting with understands you, and has relevant context about your life. Getting this right requires us to be constantly thoughtful about how we’re communicating at every moment in our conversations. It’s a fascinating, exciting design challenge that would be a lot harder to get right without adequate research.
[VG]: Talking about content, again when I look at the Design Genome project at Capital One, by InVision, it talks about design culture, a design seat at the executive table, sprints and epics, and shipping products. Even if I try to find the word “Content”, I cannot find a single instance in this report. And I strongly feel that this report misses out on talking about “content”. How would you add the “content orientation” to this report?
[LL]: Even if it isn’t mentioned explicitly in this report, much of what is talked about does include content design as a core part of the design process, present at the highest level.
At Capital One, content design means using language to create meaningful, personal interactions with customers across all touchpoints of an experience. Content designers use words to drive emotion, knowing that our choices can make or break the entire user experience if we’re not careful. Essentially, we are experience designers. It’s just that instead of visual elements, our main tool is words.
Thinking about communication FIRST should drive the end-to-end design process. This isn’t work that ONLY content designers are responsible for, however; it’s a mindset that all creators of an experience need to embrace to achieve the most successful outcome.
When content designers work with design, product, and tech teams at Capital One, we ideally start our work in a Word doc (yes, you read that right–a WORD doc). We do this to orient our cross-functional team around the conversation we want to be having with our customers throughout the end-to-end experience, and then together we begin to craft how that back-and-forth might go. By talking through our designs this way, it helps us get at the core of what matters most to our customer in each moment, and forces us to make sure that we’re communicating what they need to know in a way that makes them feel confident.
By focusing on the words we’re using right out the gate, we can ensure that we’re answering the right customer questions and providing the right information when they need it most, before anyone starts prototyping or creating higher-fidelity designs. We can test low-fidelity prototypes that literally just consist of words on a page (A WORD DOC!) with customers, to make sure that the direction we’re heading in is solving the right customer need. This approach enables us to solve for understandability at the outset, which is critical. THEN we can continue to build a beautiful interface (in fewer iterations!) that we know will not only be visually well-designed, but will actually make sense for the people using it.
Ultimately, anyone involved in creating an experience should be an expert in knowing our customers’ needs. And if we want to get to know our customers, connect with them, and solve for these needs in an understandable way, we all have to be masters of the conversation we are having with them.
When done right, these thoughtfully designed conversations help lower call-volume costs, increase conversions and customer satisfaction, and, most importantly, build trust with our customers.
[VG]: I cannot agree more on the ‘solve for understandability’ part. In an interview at Fast Company, Steph Hay mentioned the names of some roles in the design team — the AI Writer, Narrative Designer, Language Scientist. It is no brainer that AI and data must be central to lot of design process points. How does EMPATHY fits into this technical-sounding work culture? For instance, does EMPATHY checkpoints contribute to data decisions, OR you use data to identify EMPATHY opportunity in the ‘design plus content’ picture?
[LL]: What a great question!
Empathy has EVERYTHING to do with how our team shows up, both in the experiences we design for our customers and in the interactions we have with each other as the people responsible for creating these experiences.
You’re right in that these job titles do signify a bit of a shift in the role of a designer. As Steph detailed in the Fast Company interview you note above, an AI Writer, Narrative Designer, or Language Scientist is no longer operating at purely the interface or touchpoint levels when it comes to creating these experiences. We’re now working at the systems level, because the AI is going to be making decisions that inform the customer experience. While that might feel more technical at the outset, this work is still deeply creative and emotional.
It’s our responsibility now more than ever to understand the emotional contexts people are in when they are trying to do something, and make sure that this contextual data is built into the system in a way that enables more meaningful conversations between Eno and our customers.
It’s how we get to a point where we’re designing conversations that embrace the natural language that customers use to communicate with us, and helps us build experiences that get smarter with each interaction.
By taking these things into account, we can start to design conversations that feel predictive, rather than reactive (“here’s info that might matter to you in this moment” vs. waiting for someone to find what they need on an app and click the button), and move beyond transactional interactions that sound robotic, to contextually relevant conversations that evoke real emotion and, hopefully, lead to relationships rooted in trust.
When we get it right, we demonstrate that we have our customers’ best interests at heart. It helps them believe that we care about them, because we do.
[VG]: Here I take a chance to see EMPATHY living in Eno.
[VG]: You talked about context and personalization, in all communication, across touch points. It is interesting to see that the way people use information to make decisions while for shopping, for travel plans, has radically changed. They are used to consume snackable, skimmable, glanceable content. In something as critical as financial transactions where accuracy and clarity in content are absolutely critical, how do you make sure that the information is comprehensive yet small enough, so that consumers do not miss out on something important. Isn’t it bit of a challenge?
[LL]: It certainly can be a design challenge, but it’s a fun one! As a general conversation design principle, Eno champions clarity above all else in its communication style. It is straightforward, and gets right to the point when providing info to customer requests.
To get even more granular, the general framework we abide by is:
“Answer the customer’s question first; ask clarifying questions and/or provide any other additional context second.” This formula helps keep the conversation focused, while also allowing for some interactivity and dynamism to the dialogue where it makes sense.
Of course, for Eno to be able to hold its own in a dynamic conversation, it needs to understand the natural language the customer is using to ask his/her questions, which requires some additional thoughtfulness on the backend. To enable us to think through the potential complexity of a use case, we create what we call Language Trees to map the flow of the narrative threads as we’re designing. These Language Trees include multiple branches for all the different directions the conversation might go, depending on customer input and context. Applying this form of structure to our work enables us to best anticipate and design for successful conversations between Eno and our customers.
[VG]: How designing a chatbot (by highly strategic use of AI) for a financial corporation is different from doing it for any other industry or audience? For instance, to ensure accuracy, compliance, context, and of course security.
[LL]: Designing for Eno is my first foray into the Conversational UI space, so I can’t speak with any specificity to how it might differ from creating a chatbot for other types of institutions. But, of course, especially because Capital One is a financial institution, accuracy, compliance, and security are key considerations that are constantly top-of-mind for us. We care deeply about the safety of our customers’ data, and know that we have a serious responsibility to protect it. We partner extremely closely with our compliance, legal, operations, product, and tech partners throughout our design process to ensure that the information Eno provides is accurate and compliant, and that we’re presenting it to our customers in a way that is accessible, understandable, and clear.
Another major differentiator for Eno is its character. That Eno represents a financial institution factors heavily into its character design, which influences how Eno communicates.
People don’t generally believe that banks are there to represent their best interests, and don’t exactly go out of their way to communicate with their financial institution, if they can help it. It’s unfortunate, because money is such an emotional but vital part of our lives.
We should feel empowered, confident, and safe when it comes to managing our finances, and the ability to have honest, straightforward conversations with the institution handling our money can go a long way toward getting us there. That’s where Eno comes in.
It’s human nature to want to interact with the people and things that we understand, that we feel understand us. It’s how we create connections, and consequently build trust. A key way of building trust between people is to consistently demonstrate the integrity of your character in every interaction you have. Turns out, this also applies to AI. Eno’s character design serves as a framework for how it shows up to our customers, consistently in every interaction. Eno has a backstory, likes and dislikes, strengths, and even some flaws, just like a human–although, if you were to ask, Eno will let you know that it’s “…a bot and proud!”
Our goal in creating Eno’s character was, and continues to be, to build a trusted, likeable personality that people understand and can relate to. Eno is a representative of Capital One, but has its own unique, thoughtful way of communicating with people that’s entirely its own.
[VG]: In some financial corporations and institutes, the “Experience Design” processes may not be so mature. If they have the budget to invest, what would you advise them to get started. This is not about the conventional content audit and analysis steps— it is more about the cultural shift and for preparedness.
[LL]: It is SO important for financial institutions to invest in what is effectively a content strategist’s superpower: simplifying the language of complex ideas. Money is such an emotional thing, as I’ve mentioned, and feeling insecurity, doubt, distrust, fear, or disempowerment around finances can be insanely demoralizing for customers. Not feeling safe or sure about how to best handle money can negatively impact every facet of life, because for better or worse, money impacts our ability to live the life we want.
Financial institutions have the power to change the dialogue between customers and their businesses for the better if they commit to hiring the right people to create human-centered experiences with measurable outcomes for customers and the business alike. When done right, effective content leads to such desirable metrics as lower call-volume costs, higher NPS, and healthier customers. AND, demonstrating this kind of care and consideration for the customer in their journey is not something that goes unnoticed.
Understandable, meaningful, tailored experiences that feel like personalized conversations foster loyalty, which can have a significant ROI for the business, especially as customers mature. Loyalty is much harder to measure in the short-term, but will prove invaluable for those companies visionary enough to play the long game.
[VG]: Ok, this is a quick one. Can you answer this question in not more than ten words — “If I apply for AI Content Designer position in Capital One, what will you look for while interviewing me (assuming that you will be interviewing)”?
(Not more than ten words.)
[LL]: You’re a skilled writer, excited about the potential of AI.
[VG]: Lauren, thank you so much for giving us a virtual tour to your workspace — in a way. This was quite fascinating — to know the whole dynamics and variables and how things shape up in a big corporation, and specifically around AI. I absolutely loved this conversation and I am sure the audience can pick their takeaways from this post.
[LL]: Thank you again for including me in the series, Vinish! This has been a lot of fun.
Previous posts in the series:
- Content Strategy in Healthcare — An Interview with Scott Abel
- Content Strategy in Non-Profits — An Interview with Jess Sand
Next post: Content Strategy in a SaaS enterprise? A university? Sports? Watch out!
