Normalized claim
Quantified impact: 10%
It handles about 1,000 inquiries per day, representing roughly 10% of customer-service traffic.
Aydem Energy deploys Azure OpenAI for an AI WhatsApp assistant to handle customer inquiries during outages
Use case typeCustomer service automationUpdated Jun 13, 2026
Aydem Energy serves around 6 million customers across five cities in Türkiye and faces call-volume spikes during seasonal disruptions and tourism peaks. The company deployed an AI assistant on WhatsApp using Azure OpenAI Service and Softtech to provide real-time outage updates and support common service tasks.
- Organization
- Aydem Energy
- Industry
- Energy & Utilities
- Location
- Turkey
- Published
- May 2026
Reported outcomes
−75%
costCost savings
10%quantified impact75%quantified impact90%quantified impact
Strategic outcomes
Speed & agilityLaunched AI assistant in three monthsCustomer experience & trustResolved most inquiries without agentsScale & capacityScaled support for peak demandCustomer experience & trustReduced repeat customer contact
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Quantified impact: 75%
About 75% of WhatsApp inquiries are resolved without live agents.
Normalized claim
Quantified impact: 90%
Aydem Energy says 90% of customers who use the assistant never call customer service again.
Normalized claim
Cost: 75% decrease
The company reported a 75% reduction in cost per conversation versus the initial POC.
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Aydem Energy
- Provider
- Microsoft
- Maturity
- Production
- Linked source
- Microsoft Customer Stories
The company deployed an AI assistant on WhatsApp using Azure OpenAI Service and Softtech to provide real-time outage updates and support common service tasks
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 3
- 1Customer Service Automation
- 2Conversational AI
- 3Call Center Augmentation
- Seasonal climate disruptions and tourism peaks created large spikes in customer-contact volume.
- Around 70% of inquiries were related to power outages, making it difficult to staff the call center quickly enough.
- The company needed a scalable way to serve customers during outages while reducing pressure on agents.
- Aydem Energy and Softtech built a generative AI assistant with Azure OpenAI Service for the WhatsApp channel.
- The assistant integrates with internal systems and CRM data to match phone numbers, provide location-specific outage updates, and support meter readings, bill checks, and compensation claims.
- The flow includes data-protection acknowledgements, escalation to human agents for urgent or angry customers, and guardrails that restrict access by geography and scenario.
- The company also tested the system for vulnerabilities and limited the knowledge base available to the assistant to reduce cost and control token usage.
Technologies
- The assistant was launched in about three months.
- It handles about 1,000 inquiries per day, representing roughly 10% of customer-service traffic.
- About 75% of WhatsApp inquiries are resolved without live agents.
- Aydem Energy says 90% of customers who use the assistant never call customer service again.
- The assistant provides workload equivalent to 10 people and can scale to about 100 in peak periods.
- The company reported a 75% reduction in cost per conversation versus the initial POC.
Architecture
WhatsApp users interact with a generative AI assistant built with Azure OpenAI Service. The assistant connects to Aydem Energy internal systems and CRM data for real-time customer-specific updates, uses location sharing to identify outages, applies scenario-based knowledge-base restrictions, and escalates complex or emotional cases to human agents.
Implementation partners1
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: May 14, 2026Publisher: Microsoft Customer StoriesEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Market context
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.