Improving NPS by 14 points with a Croatian voice assistant for UniCredit

Industry

Financial services

Use Cases

  • Frequently asked questions
  • Natural language call routing
  • Croatian

Featuring

Vedran Mrvica
Head of Contact Center,
Zagrebačka banka

Zagrebačka banka is owned by pan-European bank UniCredit and is the largest bank in Croatia, with over 1.3 million retail customers and 60,000 corporate clients.

Challenge

25% call abandonment due to long wait times and complex IVR.

Solution

The first-ever Croatian-speaking enterprise voice assistant that routes callers using natural language and answers FAQs.

Result

27% call reduction resulting in faster routing, fewer abandons and a 14 point NPS increase.


 

Challenge

UniCredit was experiencing high call volumes, exacerbated by an outdated IVR system that was negatively impacting customer experience. Customers were spending up to two minutes navigating the IVR, followed by an additional 2-4 minute wait to connect with an agent. This led to a 25% call abandonment rate and constant pressure on agents due to long queues of callers. The IVR system, over 20 years old, had become increasingly complex and difficult to manage.

Vedran Mrvica, Head of the Contact Center at Zagrebačka banka, sought a voice automation solution to modernize the IVR and alleviate the pressure of high call volumes. However, the complexity of automating service in Croatian presented a significant challenge.

Solution

After researching and meeting with vendors, Vedran found that PolyAI’s ability to communicate in Croatian was unmatched. “When we heard PolyAI’s solution in Croatian, we knew that was what we wanted for our customers. No other vendors came close to how great it sounded,” said Vedran. UniCredit’s voice assistant is now the first to be deployed in Croatian. The UniCredit team had forecast the project would take a year to deploy, but PolyAI deployed their voice assistant in just 3 months.

The voice assistant answers a range of frequently asked questions about things including card activation, card replacement, how to add a card to a digital wallet on a smartphone, and support queries for online banking. For call types that it doesn’t yet automate, it routes callers straight to the right department, asking clarifying questions when needed.

Results

14

point NPS increase

27%

call automation

83%

faster call routing

10%

reduction in abandonment rate

UniCredit’s voice assistant Mia is now automating 27% of all calls to Zagrebačka banka, routing callers 83% faster than the previous IVR and reducing call abandonment rate by 10%. These improvements have increased UniCredit’s NPS score by 14 points in the first 6 months of the deployment.

When we heard PolyAI’s solution in Croatian, we knew that was what we wanted for our customers. Our customers can now get information quickly in their local language, and I’m really proud of what we’ve created together.”

Ready to hear it for yourself?

Get a personalized demo to learn how PolyAI can help you
 drive measurable business value.

Request a demo

Request a demo