Skip to main content
Service Solutions

Case Study: Automated classification of e-mails in customer service

Improved, AI-based service quality and faster response times for the Customer Interaction Center (CIC)

Customer's industry sector: Car manufacturer

Related topics: Bosch Service Design , Customer Experience Services and Intelligent Automation

Facts and figures

Reduction in clearing process from over 5 minutes to less than 1 minute per e-mail

More than 90% of all e-mails are automatically and correctly pre-classified by text AI

Service experts concentrate exclusively on the handling of customer requests

The challenge

We operate the Customer Interaction Center (CIC) on behalf of a German car manufacturer for several of its brands. In total, we provide 18 different services by e-mail, telephone, and chat for end customers in the German market.

A major challenge is that every single e-mail has to be opened manually and has its content checked. Only then can it be assigned to the responsible service team. Before the handling of customer requests can begin, the monotonous task of e-mail pre-classification ties up important resources and delays the actual handling process.

"Our service team all agree that their new 'colleague' has become an indispensable member of the team, and our client is very satisfied with the bot. It does its job quietly in the background, leaving our service experts to get on with the task of providing personal support to their customers even better and faster than before."
Anke Schieferdecker, National Operations Manager, Bosch Service Solutions

The solution

Bosch Service Solutions began by conducting a holistic analysis of the existing processes. In the course of this review, we optimized and standardized the manual work steps in order to develop intelligent e-mail automation for incoming mail.

The solution uses AI-based text recognition based on, among other things, Natural Language Processing (NLP). Artificial intelligence is able to understand human language and interpret the context. The system also continues to learn during operation, becoming more accurate in the process. In addition, our data engineers teach the bot about new topics.

A bot now supports the customer service team with the automated e-mail clearing process:

  • Incoming mail check: The bot scans all new incoming e-mails, filters out spam messages, and opens all other relevant customer requests.
  • Identification of content: The system analyzes the subject and text and can identify the content using a predefined topic matrix and learned algorithms.
  • Changes in CRM: Ticket-relevant information from the e-mail is automatically transferred to CRM. This is the basis for classification.
  • Message allocation: The e-mail is made available in the corresponding queue.
  • Further handling by customer service: Employees who specialize in the respective service can handle customer requests directly.
  • Intelligent learning: After final handling of the request, our bot continuously improves its accuracy through machine learning (ML) based on the generated data.

The outcome

The collaboration with Bosch Service Solutions delivers numerous sustainable benefits for the Customer Interaction Center (CIC). By using text AI, more than 90% of incoming e-mails can be automatically assigned to the correct service team.

The average handling time for the clearing process per e-mail has been reduced from originally more than 5 minutes to less than 1 minute. This corresponds to 5 FTEs, who we can now fully schedule for responding to customer requests.

E-mail automation has also increased motivation among the service experts, since their job involves fewer monotonous and error-prone tasks.

Get in touch with Bosch Service Solutions

We look forward to your inquiry.


Send us an email

Do you have any questions, ideas or requests? Write to us using the contact form.


Are you interested in one of the following topics? Click on the tag and you will get all the relevant stories on this topic.