Under Rosa’s leadership, CaixaBank rolled out an IBM Watson Assistant-enabled virtual agent in 2019, designed to support the call center that handles inquiries from the bank’s 29,000 employees and 15.6 million customers. Now, CaixaBank’s multiple chatbots answer more than a million conversations each month, in Spanish and Catalan—resolving issues quickly, extending support to 24-7 and enabling continuous learning about the questions posed by clients and employees.

What was the business opportunity you sought to address by using AI?
AI has multiple applications in banking. After we began transforming the contact center experience using AI-powered virtual assistants, we began applying AI to a wide range of use cases, including process automation, text analytics and decision-making support. AI is progressing at a huge speed, and there will be a day when AI will be “software as usual”: It will be applied in practically all organizations, processes and services. Thus, a big challenge for the future is to be prepared for AI pervasiveness.

What has your business achieved with Watson?
At CaixaBank, quality improvement and cost reduction were the main benefits. For instance, with chatbots we now have 24-7 interaction, faster resolution of questions, access to all sources of documentation and, most importantly, continuous learning about the questions posed by our clients and employees.

What has your business achieved with Watson?
First, it’s really important to have the agreement and buy-in of all stakeholders. Most AI projects require important process and cultural changes. It’s not just a technology project, it is a whole company project. Second, it’s key to always keep in mind the needs of the final users and collaborate 100% of the time with the business divisions to avoid implementing something useless. Third, if possible, try to develop a scalable architecture for the solution; surely there will be more use cases that may benefit from it.

What advice would you share with others who are considering using AI?
Understanding the business case is a crucial phase in the project and can take more time than expected. Sometimes, this phase will result in a non-AI project, just a “software as usual” improvement that might be a simpler or cheaper way to solve the problem. If this is the case, do not get frustrated. There will be more AI applications for sure.

What advice would you give your younger self?
I’d emphasize the importance of being surrounded by the best professionals. You will learn from them, and if you can have them as mentors, they will really help you progress in your career. I’d also emphasize the necessity of being proactive and not despairing. There is plenty of time for everything, and where there’s a will, there’s a way.

She puts Watson Assistant to work

Dive deeper into your journey to AI and learn more about Watson Assistant