A European energy utility
71% of customer cases resolved autonomously, in two languages, inside the utility’s app.
A utility embedded an Agentforce agent in its mobile app for customers with a contract or a smart meter. It combines CRM data with the company’s own knowledge base to explain a customer’s situation in their language, and it can take action, such as switching on smart charging when a customer mentions an electric car.
At a glance
- Client
- A European energy utility
- Engagement
- A grounded, bilingual agent embedded in the customer app
- Sector
- Energy & Utilities
- Based in
- Belgium
The stack
01
The situation
What they were working against.
A utility fields high volumes of routine questions about billing, contracts, pricing, smart meters and EV charging, and has to answer them consistently across channels and in two languages. It wanted 24/7 self-service that could explain a specific customer’s situation rather than recite policy.
02
What was built
The build, in the order it happened.
Grounded on the customer and the knowledge base
Agentforce combining Salesforce CRM data with a custom vector knowledge base, so the answer is about this bill and this meter.
In the app, in the language
Deployed inside the customer mobile app for anyone with a contract or smart meter, answering in either of the country’s two main languages.
Actions, not just answers
When a customer mentions owning an EV, the agent can activate smart-charging options in the app rather than telling them where to click.
03
The results
Most questions never reach a person, and the ones that do arrive explained.
The agent handles more than 1,500 conversations a day and resolves 71% of cases autonomously, with web and email channels planned next.
- Cases resolved autonomously by the agent
- 71%
- Conversations handled every day
- 1,500+
- Languages served from one agent
- 2
Client name withheld. Figures are as reported for this implementation; where one was a stated target rather than a measurement, the label says so.
04
Our read
How we would build this at your scale.
What we take from this build, and what the same outcome looks like for a team of fifty to five hundred people rather than a global enterprise.
The knowledge base decides the resolution rate
Seventy-one percent is a function of how much of the customer’s situation the agent can see and explain. We ground on billing and meter data through the CRM, not on a PDF of the tariff.
Agentforce & AI 02An agent that can act needs scoped actions
Turning on smart charging is a write to a live system. We define each action, its guardrails and its audit trail as an integration contract before it is exposed.
Integrations 03Embed where the customer already is
For a regional utility that means the app or the portal. Building the surface in Experience Cloud or a native app is a development job we scope alongside the agent.
Salesforce DevelopmentMore case studies
One of the largest U.S. homebuilders
An after-hours sales agent that helps up to 30% of inbound leads book a home tour.
ReadA private university with an ambitious enrolment target
An admissions agent built to take enrolment from 10,000 to 50,000 students in five years.
ReadA national product-philanthropy nonprofit
Disaster-relief donations routed up to three times faster by a resource-matching agent.
ReadWant the same result, without the enterprise bill?
Thirty minutes with an architect. Bring the part of this story that looks like your problem and we will tell you what it takes to get there: which products, how many weeks, and what we would leave out.