AI in practice: Six business cases across the product lifecycle

Press Releases

GenAI offers great potential for process automation in particular. In the new "GenAI Impact Report 2025" from adesso, half of the 400 managers surveyed in Germany stated that they could automate more than 30 per cent of their everyday tasks with the help of GenAI - across all areas of business.

The IT service provider describes specific use cases to show how AI can be strategically implemented along the various phases of product lifecycles to optimise processes, accelerate innovation, and improve customer experience:

1. AI-controlled engineering in product and process optimisation:

In modern production environments, networks of sensors continuously capture data that are analysed in real time by machine learning algorithms. These technologies detect anomalies in manufacturing processes and support predictive maintenance concepts. GenAI algorithms make it possible to generate thousands of design variants and test them in simulated load scenarios. This AI-supported process optimisation allowed one automotive supplier to reduce the development time of new components by around 30 per cent and react flexibly to changing production conditions through adaptive engineering.

2. Personalised marketing through AI technologies:

Modern marketing automation uses a variety of AI technologies to generate and manage highly personalised marketing campaigns across all contact channels. Large Language Models (LLMs) analyse extensive volumes of customer data and use GAN (Generative Adversarial Networks) to create dynamic content for social media, emails, and websites. In customer contact, chatbots powered by NLP (Natural Language Processing) take on personalised interaction roles and achieve significantly higher conversion rates. One e-commerce platform was able to increase its click-through rates by more than 25 per cent using these technologies. At the same time, more precise targeting significantly reduces advertising costs. Continuous customer sentiment monitoring also enables hyper-personalisation in real time. However, companies must be careful to achieve a balance between personalisation and data protection.

3. Predictive selling in sales:

In sales, GenAI-supported CRM systems enable accurate predictions about optimum customer offers. Graph Neural Networks (GNNs) analyse the complex relationships between products, customer interactions, and sales transactions in order to identify patterns. Bayesian Inference is used to calculate the probability of customer reactions to specific offers. Reinforcement Learning (RL) tests different sales strategies and optimises Next Best Actions (NBA) in real time. By combining these technologies, one B2B company was able to increase its completion rates by 40 per cent. However, successful implementation requires not only technical expertise, but also change management and further training for salespersons.

4. Automatic pattern recognition and fraud detection in transaction processing:

AI-supported analysis systems improve the detection of irregularities in transactions. Anomaly Detection with Unsupervised Learning identifies atypical patterns in transaction data. Computer vision technologies automatically validate submitted documents. A financial services provider was able to detect cases of fraud 70 per cent more quickly and at the same time optimise the prioritisation of support requests using automated clustering models. It is important to continuously develop the models in order to adapt to new fraud scenarios and to ensure human oversight when critical decisions are being taken.

5. AI-supported compliance automation and fraud detection in risk management:

In the insurance and finance sector, NLP technologies and knowledge graphs are used to analyse contract documents and loss reports. Transformer models classify suspicious transactions with high precision. Federated learning enables companies to train AI models together without directly exchanging sensitive customer data. In practice, this AI-supported risk analysis has reduced the manual review effort by around 50 per cent, while legitimate claims are processed more quickly. Human expertise remains indispensable for critical cases.

6. Smart automation in customer management:

The integration of conversational AI into customer service processes allows for more natural dialogue and the handling of complex enquiries. Speech-to-text models analyse calls in real time and identify customer needs. Sentiment analysis algorithms record the customer's mood and adapt the answers accordingly. After implementing such systems, one telecommunications provider was able to automate more than 75 per cent of routine enquiries, while simultaneously significantly increasing customer satisfaction through personalised, context-dependent communication. However, companies must handle AI-based communication transparently and always ensure a smooth transition to human employees.

Press service

Materials for media reports

With our press service, we provide you with various materials that complement our corporate news or serve as background information.

To the press service

Do you have any questions?

Do you have any questions about our press service or a specific request? Please contact us.

Communications Manager Nils Roos +491704406148 nils.roos@adesso.de