20. February 2025 By Sandra Weis
Semantic analysis with AI: efficient evaluation of customer surveys in the bAV
Occupational retirement provision (ORP) is a central component of modern human resources strategies. For companies, it offers a competitive advantage in employee retention, while providing financial security in retirement for employees. But insurance companies, pension providers and financial service providers also recognise the enormous sales potential of occupational pension schemes. To ensure their long-term success, a high level of customer satisfaction is essential. Customer surveys play a crucial role here: they help to identify potential for improvement and to address the needs of different target groups in a targeted manner – be they employers, employees or sales partners.
The challenge: unstructured data from customer surveys
Customer surveys often yield large amounts of unstructured data. Open-ended text responses in particular contain valuable information, but manual evaluation is extremely time-consuming and prone to error. Instead of spending hours or even days reviewing and interpreting texts, semantic analysis using AI can analyse complex data sets in no time. It processes this data quickly and efficiently, understands connections and enables in-depth insights into the wishes and expectations of the customer base. In this way, the wishes and needs of the customer base are not only made visible, but also more comprehensible – a valuable basis for implementing targeted improvements and making better decisions in the long term.
Rethink company pension schemes – with adesso!
Design the future of company pension schemes to be digital, efficient and legally compliant. With customised IT solutions from adesso, you can optimise processes, increase user-friendliness and secure long-term competitive advantages. Let's drive innovation together – for a company pension scheme that pays off!
Find out more about our digital services for company pension schemes
Why are customer surveys so important in the occupational pension system?
The occupational pension system is aimed at different target groups, each with their own specific requirements:
- 1. Employers – as contractual partners of the occupational pension system, they are looking for transparent and uncomplicated solutions.
- 2. Employees – they want clear information and easy access to services.
- 3. Sales partners – for them, clear products and optimised service processes are essential to successfully broker the bAV.
Customer surveys help to understand the perspectives of these groups and to address them in a targeted manner. Typical questions are:
- How satisfied are you with the bAV products and services?
- What hurdles do you experience when taking out or using them?
- How do you rate the service and communication?
- What improvements would you like to see?
Free text responses in particular often contain detailed suggestions, criticisms or requests that provide valuable information on potential improvements. However, manually analysing this data is very time-consuming and prone to error.
What is a semantic analysis?
Semantic analysis using AI makes it possible to comprehensively understand and systematically evaluate texts and data. In the process, the content is interpreted and processed in the correct context. The aim of semantic analysis is to understand the meaning of texts by analysing the relationships between words, phrases and sentences, as well as recognising patterns and anomalies in the responses and extracting relevant information based on these. For example, the AI can analyse whether a comment is positive, neutral or negative. It can also identify recurring criticisms, such as frequent complaints about ‘lack of transparency’. Furthermore, the AI can create topic clusters by automatically assigning similar statements to specific topics. This makes it possible to identify which topics are addressed particularly frequently. The results of these analyses provide a basis for targeted improvement measures. Insurance companies or pension providers can thus make well-informed decisions to optimise processes, communication or products.
The advantages of semantic analysis
AI-supported semantic analysis offers numerous advantages:
1. Efficiency in evaluation
A conventional manual approach to analysing free text responses is time-consuming and prone to error. AI models can analyse responses and prepare them in a structured way in a short period of time. This saves time and significantly reduces effort compared to manual evaluation.
2. Objectivity
In contrast to manual analysis, AI is free of unconscious biases. AI models work neutrally and consistently, which makes the analysis more reliable.
3. Detection of hidden patterns
AI models identify patterns that are often not obvious to human analysts. For example, criticisms of the ‘complexity’ of an occupational pension solution could be repeatedly linked to negative comments about ‘communication’.
4. Customisation
By analysing feedback from specific employee or employer groups or from sales, tailored solutions can be developed that address specific needs.
Conclusion
Semantic analysis with AI opens up completely new possibilities for occupational pension schemes. It helps to better understand the needs and opinions of target groups and to respond to them in a targeted manner.
By analysing customer feedback in a targeted manner, providers of corporate pension schemes can efficiently improve their products, services and communication processes. Insurance companies and pension providers that use AI-supported text analysis show that they actively seek out customer feedback and take their customers' opinions seriously. This builds trust and gives them a clear competitive advantage in a dynamic market.
adesso is your partner for the successful implementation of innovative ideas in the field of company pension schemes, such as semantic analysis with AI. Benefit from our consultants‘ and developers’ industry knowledge and expertise in the company pension scheme environment – from the introduction of digital solutions across all implementation options.
