AI-Powered Performance Discovery
Automated Quality Management identified the agents with the highest retention rates by analyzing conversational behaviors across thousands of interactions, using evidence rather than reputation.
Customer success story
How MAPFRE increased retention from 1% to 40% and protected BRL $1M of revenue in just two months using AI-powered Quality Management
Learn how insurance giant identified revenue loss and mobilized their best agents to retain customers.
1% → 40%
Customer retention rate on cancellation calls
BRL $1MAbout USD $200k
Revenue protected
2 months
From rollout to measured results
75k
Calls a month analyzed, not sampled
MAPFRE Brazil
Summary
Retention is one of the few levers a contact center controls directly. Most quality programs can’t reach it, because they review a sample. The behaviors that save a customer sit in the calls nobody listened to.
MAPFRE Brazil had a specific version of that problem. Customers asking to cancel a policy were routed to a general service team with no retention strategy. Most requests were simply processed, so customers churned and recurring revenue went with them.
Using SuccessKPI Automated Quality Management, MAPFRE turned that cancellation process into a retention operation built on data. Within two months, the company had identified its strongest retention agents, formed a specialized team, standardized what those agents did well, and moved the numbers.
“SuccessKPI GenAI enabled us to discover what traditional quality monitoring could never reveal. By identifying our highest-performing retention behaviors and transforming those insights into a specialized operation, we dramatically improved customer retention and protected significant business revenue in only two months.”
Celso Roberto TeixeiraQuality Manager, MAPFRE
The problem
Business Challenge
Before the implementation, every customer selecting the cancel option in the IVR went to the standard customer service team. Those agents were experienced in support work. They weren’t trained or organized to hold onto a customer who had already decided to leave.
The organization couldn’t answer four questions that mattered. Which agents consistently retained customers? What conversation techniques did they use? How could those behaviors spread to everyone else? And which interactions carried the most value to the business?
Manual quality monitoring reviewed a limited number of calls. Listening to a handful per agent each month won’t surface a behavioral pattern across 75,000 calls. Revenue was walking out every day through a queue nobody was watching closely.
What changed
The Turning Point
Rather than adding headcount or redesigning the IVR, MAPFRE used AI to understand what was said on the calls. SuccessKPI analyzed thousands of cancellation conversations at full volume instead of a sample. AI Auto-Scoring answered the evaluation questions the quality team configured, and an analyst applied or overrode each suggested answer.
A pattern showed up quickly. A small group of agents held onto customers far more often than everyone else. They negotiated differently, handled objections earlier, showed more empathy, and moved the conversation to the value of the policy rather than the price of leaving.
For the first time, MAPFRE could describe what great retention sounded like, in the agents’ own words, and could point to the calls that proved it.
What MAPFRE built
Solution
Creation of a Dedicated Retention Team
The highest-performing agents were consolidated into a specialized customer retention team, focused exclusively on cancellation requests and the revenue at risk inside them.
Standardization of Best Practices
The most successful conversations were analyzed to identify the techniques that consistently influenced customers to stay. Those insights became the foundation for coaching, training, and operational standards.
Intelligent Customer Routing
When a customer selects the cancellation option in the IVR, the interaction now goes straight to the retention team. High-risk moments reach the people who handle them best.
Results in two months
Business Impact
| Measure | Before | After |
|---|---|---|
| Retention rate on cancellation calls | 1% | 40% ↑ |
| Quality coverage | Sample-based manual review | Every call in scope scored on the configured questions, with human review |
| Cancellation handling | General service queue, no retention plan | Dedicated retention team, routed from the IVR |
| Coaching input | Evaluator judgment on a few calls | Techniques drawn from the conversations that actually saved customers |
| Revenue protected | Not measured | More than BRL 1 million, about USD 200,000 |
What else moved
Beyond Financial Results
The project delivered measurable improvements across multiple business dimensions.
Customer Experience
- More personalized conversations, tailored to why the customer wants to leave
- Better objection handling, by people who handle those objections daily
- Increased customer confidence at a moment that decides the relationship
- Improved service experience during critical customer moments
Employee Performance
- Top performers identified from their own calls, not from anecdote
- Data-driven coaching and development, anchored to real conversations
- Higher employee engagement on the work that carries the most value
- Greater recognition of high performers, and a path to spread the skill
The difference
Why SuccessKPI Made the Difference
A quality leader wrote the prompt and tested it on a real call
Per-question guidance prompts let MAPFRE encode judgment instead of keywords. Every suggested answer carries its reasoning, so a supervisor can read why a call scored the way it did.
A person stayed on every score
The evaluator applies or overrides each suggested answer. Coverage rose without asking anyone to trust a number they couldn’t check.
One scorecard covered the whole program
Auto-scoring runs against the same evaluation forms human evaluators already use. One rubric, one dispute path, and one record of how a score changed.
The finding became an operation
MAPFRE turned the pattern into a team, a routing rule, and a coaching standard. Measuring agent performance was the start. Changing what the operation does with that measurement is where the 40% came from.
What it took
Key Success Factors
- Automated Quality Management with AI scoring
- Analysis across the cancellation calls in scope
- Behavioral intelligence at scale
- Data-driven identification of top performers
- Creation of a specialized retention team
- Continuous coaching based on real customer interactions
- Revenue protection through artificial intelligence
The view from the top
Executive Perspective
- MAPFRE did not improve retention by hiring more agents.
- It did not launch aggressive discount campaigns.
- It did not redesign its IVR.
Instead, the company used the intelligence sitting inside thousands of customer conversations. By combining AI, automated quality monitoring, and operational discipline, MAPFRE turned a reactive cancellation process into a deliberate revenue protection strategy.
In two months, a traditional customer service operation became a retention organization that delivers measurable business value.
What are your cancellation calls telling you?
If quality review covers a sample, the behaviors that save customers are sitting in calls nobody has heard. We’ll walk your team through how scoring and coverage work on your own platform and your own calls.