General briefing

Wilken Software Group is a leading software provider specializing in business solutions for various industries, with a significant focus on municipal utilities and energy providers. Their software suite supports the entire business process chain of utility companies, from meter reading and billing to customer relationship management.

Municipal utilities and energy providers face complex challenges in customer communication, particularly regarding invoice comprehension. As a technology partner for these companies, Wilken aims to enhance their software solutions with AI capabilities to address these challenges and improve the overall customer experience.

Challenge

Annual utility bills, especially for electricity, contain numerous complex details including consumption values, tariff calculations, taxes, and various fees. For end customers, these invoices are often difficult to understand, leading to confusion and a high volume of inquiries. This creates a significant burden for customer service departments at municipal utilities, who must allocate substantial resources to clarify invoice contents.

Customer service departments at municipal utilities currently handle thousands of billing-related inquiries annually, creating operational inefficiencies
Customer satisfaction is negatively impacted when billing information is unclear
The complexity of energy billing is increasing due to regulatory changes and the energy transition
Efficient customer communication is a competitive advantage in the liberalized energy market

In addition, muncipal utilizies and enegery provides face growing skilled labor shortage. This means customer touchpoints cannot be adequately staffed to handle peak inquiry volumes, a problem expected to worsen in coming years Further, energy providers are actively pursuing a “channel shift” strategy, transitioning customers from traditional customer service to more efficient self-service options.

Regarding the channels, today’s utilities operate multiple customer communication channels including customer portals, mobile apps, telephone hotlines, physical service centers, chat systems, and email support. While billing inquiries can originate through any of these channels, human representatives are typically required to provide explanations across all platforms. Implementing AI agent support across more channels would create significant value for both utility providers and their end customers

Considering these challenges, Wilken raises following question:

How can agentic AI be leveraged to develop an intelligent solution that reduces burden on customer service departments and increases customer satisfaction?

Mandatory questions

Following questions should definitely be answered as part of your concept and adressed by your MVP

What are unmet needs of parties involved in the process?
How does your solution address the key requirements and unmet needs?
What data is available and can be used?
What capabilities do AI agents bring in and how can these effectively used?
How should domain knowledge about utility billing (tariffs, regulations, taxes, fees) be represented in a system to enable accurate and contextual explanations?
How can we ensure the solution complies with relevant data protection regulations while handling sensitive customer billing information?
Which open-source components are suitable for an MIT/Apache-2.0-licensed system?
What is the potential impact for Wilken’s customers (i.e., business value) of your solution?
How can Wilken monetize the solution?
How does the MVP proof your concept, particularly regarding technical feasability and value add?
What factors must be taken into consideration to ensure the successful implementation and ongoing maintenance of the solution?
How can explanation quality be continously be measured and improved?

Requirements

Measure of impact: It is essential to introduce metrics that allow to evaluate the effectiveness and performance of the solution.
Interactive user interface: The solutions should allow customers to explore their invoice with guided explanations.
Domain knowledge: The solution must draw on soild domain knowledge about utility billing and allow to maintain that knowledge.
Performance & accuracy: The solution should achieve high accuracy in invoice interpretation and explanation generation, while process and respond to customer queries in real-time.
Security and data privacy: The solution should implement appropriate security measures for handling sensitive customer data
Technology foundation: The solutiuon should utilize open-source technologies for core AI/ML components and follow modern software architecture principles (API-first, microservices)

Goals and outcome

The goal of this challenge is to develop an innovative AI-based solution that transforms how utility customers understand their invoices, while significantly reducing the support burden for municipal utilities. A successful solution will:

Empower customers with clear, accessible explanations of their utility bills
Drastically reduce the volume of invoice-related support inquiries
Enhance customer satisfaction and loyalty for municipal utilities
Demonstrate the value of AI in improving customer communication

The winning team will have the opportunity to present their solution at Wilken headquarters and discuss potential implementation with Wilken’s innovation and/or product development team.

Knowledge base

Details about Wilken an their services can be found online on their Corporate Website
Sample utility invoices in PDF and XML formats
General information sources on utility billing:
- Static invoice explanation tool by EnBW
- Invoice Information as provided by the Federal Network Agency
- Federal Ministry for Economic Affairs and Climate Action - Electricity Price Components
Documentation on common support inquiries related to utility invoices (see below)

In addition, Christina Plugge and Dennis Lettner are available for Q&A sessions at following time-slots:

Wed, April 16, 10 – 11 am (via Teams, link see Moodle)
Tue, May 6, 10 – 11 am (via Teams, link see Moodle)

Examples of Common Customer Support Inquiries

To better understand the types of questions the AI solution needs to address, here are common examples from municipal utilities:

Questions about invoice atructure (Navigation Help)

Where can I find specific information on my invoice?

Questions about cunsumption clarification

How was my consumption determined?
Why was my consumption estimated (possibly despite having reported a meter reading on December 26th - background: accounting reference date is December 31st)?
How does the estimation process work?
Why do I have different consumption time periods on my invoice?
How are the consumptions per time period calculated? How is it weighted?
For gas: How does the conversion of my meter reading (m³) to kWh (calorific value/altitude zone) work?

Questions about pricing:

What is the working price? What is the basic price?
How was my new customer bonus applied? (15% - possibly only on the working price?)
Why has my price changed? (communicated price adjustment/increase)
The price on my invoice is different from the one on my contract confirmation - why? (Purchase of the price before notification of levy changes)

Questions about cash flow:

Why weren’t all my installment payments taken into account on the invoice? (Payment receipt vs. time of invoicing)
Why do I have to pay an additional amount, even though I consumed less compared to the previous year? (e.g., price increase or “12th installment” is collected via invoice)
I believe I paid more installments?
Why is a late payment fee being charged - I believe I have paid?
How is my remaining amount to be paid calculated? (Possibly an offset with the first installment for the new accounting period takes place here)
Why is my next installment amount increasing, although my consumption has remained almost constant? (Possibly changes to the price - such as a tax increase - are already taken into account in the installment calculation)