E.ON
multichannel conversational agents (voice + chat)
E.ON runs more than 30 Cognigy.AI conversational agents that automate 70% of over two million customer conversations per year, across phone and chat, in Germany and the Netherlands.
Key points
- Multichannel voice and chat conversational agents for routine customer service tasks.
- Cognigy.AI platform, more than 30 agents in production, including Robin at Essent.
- 70% automation across more than 2 million conversations per year.
- Evidence level B, confirmed active status.
Objective
Automate repetitive requests across every channel to reduce contact center load and cost, while keeping advisors available for high-value cases.
The deployment
E.ON runs a portfolio of more than 30 conversational agents built on Cognigy.AI, active over phone and chat around the clock. They handle routine tasks: meter readings, self-service billing, end-to-end contract changes, and payment detail updates. When a request falls outside their scope, a handoff to an advisor is in place. The Dutch subsidiary Essent built an award-winning agent, Robin, on the same foundation, also operating over phone and chat. Together they handle more than two million conversations per year, with a 70% automation rate across customers and staff.
Results Proof B
Quantified platform case study (Cognigy) with annual volumes and automation rate, echoed on the vendor's customer page. Sources agree but both come from the vendor, hence a level B rather than higher.
How it works
Documented architectureThe stack in detail
- plateforme Cognigy.AI Conversational AI platform that runs more than 30 voice and chat agents; 70% automation across more than 2 million conversations per year.
- outil Connecteurs back-office E.ON Access to account, billing, and contract systems to run meter readings, contract changes, and payment details end to end.
- outil Agent Robin (Essent) Award-winning voice and chat agent from the Dutch subsidiary, built on the same Cognigy.AI foundation.
- infra Canaux telephone et chat E.ON Around-the-clock coverage on both channels, with a conditional handoff to an advisor.
How it runs, concretely
For ops teams-
1Multichannel intake AI
The customer arrives by phone or chat; the agent identifies the intent and the account context.
-
2Task execution AI
The agent completes the routine operation (meter reading, billing, contract change, payment details) end to end.
-
3Conditional handoff AI
Out of scope or complex cases, the agent hands off to an advisor with the context.
-
4Continuous improvement data team
Teams analyze aggregated conversations to adjust the flows and the service portfolio.
The intent detected in the conversation and the resolution rate without a human. If intent is misclassified, handoffs to advisors rise and the automation rate drops.
How your customers perceive this type of use
Sourced studiesLes consommateurs n'acceptent pas les chatbots par defaut : 64% prefereraient que les entreprises n'utilisent pas d'IA dans leur service client (Gartner, 2024) et pres d'un utilisateur sur cinq du service client par IA n'en retire aucun benefice (Qualtrics, 2025). L'acceptation se construit sur trois conditions mesurees par Salesforce : savoir qu'on parle a une IA, pouvoir escalader vers un humain, comprendre la logique de l'agent.
Acceptance conditions
- Etre informe qu'on parle a une IA et non a un humain (pres de 75% le demandent, Salesforce 2024)
- Un chemin d'escalade clair vers un agent humain (45% plus enclins a utiliser l'agent IA, Salesforce 2024)
- Une logique de l'agent clairement expliquee (44% plus enclins, Salesforce 2024)
Red lines
- Rendre l'humain injoignable : c'est la premiere inquietude des consommateurs sur l'IA dans le service client (Gartner 2024) et 50% craignent que l'IA les coupe du contact humain (Qualtrics 2025)
- Remplacer le service client par l'IA sans alternative : 53% envisageraient de partir chez un concurrent (Gartner 2024)
Sources: Salesforce 2024 · Gartner 2024 · Qualtrics 2025
How to replicate
Inference, not sourcedData prerequisites
- catalog of intents and self-service tasks
- API access to billing and contract systems
- conversation history for training
Org prerequisites
- business teams able to design and maintain the flows
- governance for handoffs to advisors
Possible stack
- conversational AI platform (Cognigy, Kore.ai, or equivalent)
- CRM and billing connectors
- voice + chat engine
The plan, step by step
- Step 1Pick two or three high-volume, low-complexity tasks: meter reading, invoice date.Deliverable: Prioritized self-service scope with volumes
- Step 2Build the flows on the platform and the API connectors to billing and contract.Deliverable: End-to-end flows in a test environment
- Step 3Launch on chat and measure the automation rate.Deliverable: First agent in production with a dashboard
- Step 4Add the voice channel on the same tasks.Deliverable: Voice agent in production
- Step 5Expand the task portfolio and set the governance for handoffs to advisors.Deliverable: Agent portfolio and documented handoff rules
First step: Pick two or three high-volume, low-complexity tasks (meter reading, invoice date) and automate them on chat before adding voice.
The same pattern, proven in other industries
ADT
genAI customer agent
Air India
genAI customer agent (automated customer service)
AT&T
Governed enterprise genAI assistant: an LLM connected to internal knowledge, open to tens of thousands of employees and called at scale by automated systems
Sources
- S1 E.ON's AI Agents Provide Best-in-Class Service Interested party archive pending
- S2 Customer Success Stories - Cognigy Interested party archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.