A national mobile, fibre and TV operator
Answers for 6,000 contact-centre reps in 30 seconds instead of minutes, at 95% accuracy.
A national operator serves millions of mobile, fibre and TV customers through a distributed contact centre. Its agent sits in the Service Cloud console reps already use and answers from the knowledge base and the customer’s live data, so a customer who used to wait minutes for an answer now waits about thirty seconds.
At a glance
- Client
- A national mobile, fibre and TV operator
- Engagement
- Data 360 and an agent inside the service console
- Sector
- Communications
- Based in
- France
The stack
01
The situation
What they were working against.
Reps in a contact centre of employees and outsourced partners had to search and cross-reference hundreds of knowledge articles by hand. Customers waited minutes for an answer and consistency varied from call to call.
02
What was built
The build, in the order it happened.
The data underneath
Data 360 unifying CRM records, case history, satisfaction scores and technician history with data from the ERP, a cloud data platform and an on-premises data lake.
An agent in the console
An employee-facing Agentforce agent embedded in the service console, returning summarised answers from more than 500 articles and live service data, surfacing pre-call context and drafting follow-up texts and emails.
Out to the field
The same guidance extended to 10,000 field technicians through Field Service, inside the mobile app they already carry.
03
The results
Minutes to seconds, for six thousand people at once.
Reps get accurate answers in about 30 seconds instead of minutes, and the agent reached 95% accuracy. The field rollout is designed to unlock 10,000 more service appointments a day without growing the workforce; that figure is the design goal rather than a measured result.
- Contact-centre reps supported by the agent
- 6,000
- Answer accuracy
- 95%
- Extra service appointments a day the field rollout is designed to unlock (target)
- 10,000
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.
Start with an employee-facing agent
The agent helps a rep, who still talks to the customer. That is the lowest-risk, highest-adoption place to begin, and it is where we point most service organisations first.
Agentforce & AI 02Accuracy is a knowledge-base project
Ninety-five percent comes from 500 articles that were worth reading. We audit and restructure the knowledge base before grounding anything on it.
Salesforce Consulting 03The integrations are the hard part
An ERP, a cloud data platform and an on-prem lake feeding one profile is a MuleSoft-shaped problem, sized down but not simplified for a regional operator.
MuleSoft IntegrationMore case studies
A global motorsport championship
Race-day issues resolved 80% faster, with one profile for 24 million fans.
ReadA European energy utility
71% of customer cases resolved autonomously, in two languages, inside the utility’s app.
ReadOne of the largest U.S. homebuilders
An after-hours sales agent that helps up to 30% of inbound leads book a home tour.
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.