Case study – Santander Bank

Case study – Santander Bank

Santander Bank is the third largest bank in Poland in terms of the value of assets and third in terms of the number of branches. In 2018, the Bank went through a merger with Deutsche Bank - on November 10, Deutsche Bank's customers became Santander Bank's customers.

Such a complex merger and migration of customers brought many challenges in customer service departments, where almost 400 consultants had to quickly expand the range of supported products and services with a completely new group of customers. On the other hand, the bank's clients had to struggle with the extended waiting time for the correct response from consultants via chat, audio, video, e-mail and the contact form.

Due to the merger, the need to quickly search a broad and complex knowledge base has increased even more.


  • Extended response time to customers
  • Introduction of a new employee to the customer service department was cost- and time-consuming - the new employee was ready to take up the position only after a month of preparatory training, and then worked under the supervision of an experienced consultant for up to 6 months
  • High employee turnover in the customer service department
  • Large division and thematic specialization of teams
  • A significant percentage of inquiries required consultation with other, more experienced consultants
  • Lack of consultation or confirmation of information increased the risk of giving an incorrect answer, which in turn resulted in complaints


Consultants decided to use BOTWISE as a platform to integrate numerous documents in one place and search them using simple natural language queries - they could connect external websites, resources stored on the intranet and network drives. For the question asked, they were given a part of the document, along with the original document.

What is important, they could also get started quickly with the operations, because all document preprocessing happens automatically - there is no need to manually rework documents, tag or build conversation scenarios.

In addition, they could analyze inquiries and extract valuable information, e.g. regarding:

  • the most popular topics
  • the number of questions asked by specific groups of consultants
  • periods of heavy traffic
  • questions which customers ask and which had not been answered in the knowledge base yet
Santander bank - usage of BOTWISE solution graph


  • 97.6% - correct answers after a month of working with BOTWISE
  • 7% - this is how much time of consultants' conversations was reduced
  • 89.7% - the level of satisfaction among users of the BOTWISE application
  • the onboarding time for new employees was shortened
  • the number of complaints has decreased
  • the percentage of calls requiring the involvement of the second line of support has decreased
The solution was implemented and used by approx. 300 consultants. This allowed us to reduce the number of advisers' inquiries to the Bank support line in a strategic moment, thanks to which the time of conversations was reduced by 7%. In the analyzed period, 8,827 queries were sent to the chatbot, of which 97.6% of the intentions were recognized.

Ilona Dawidowska, Director of the Development and Support Office, Multichannel Communication Center