What challenges did Fractal Analytics face before AI implementation?

Knowledge workers and call center agents needed faster access to accurate information across large, unstructured document repositories. The company wanted to reduce average call handling time, lower call deflection gaps, and minimize compliance risk from outdated answers.

What was the impact of the AI solution?

The system handled hundreds of thousands of queries per month across more than 10,000 documents. The client observed a 10-15% reduction in average data retrieval time and around a 30% call deflection rate.

Source

PilotEvidence: Medium65/100

Fractal Analytics reduces call handling time by up to 15% with Amazon Bedrock knowledge assist

Use case typeKnowledge managementUpdated Jun 13, 2026

Fractal Analytics built Knowledge Assist as a unified knowledge base for enterprise knowledge workers and contact center agents. The solution is designed to speed up retrieval across large, unstructured internal document sets while improving accuracy and reducing compliance risk from outdated information.

Organization: Fractal Analytics
Industry: Professional Services
Location: India
Published: May 2026

Reported outcomes

Time: −10–15%

Time & speed

Catalog median for time & speed deployments: −50% across 295 reported metrics. Compare benchmarks →

Planned next steps

During the pilot, nearly 500 knowledge workers adopted the solution.

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Time: 10-15% decrease

AWS case studyMay 13, 2026Case studyInferred claimMedium evidence strength

The client observed a 10-15% reduction in average data retrieval time and around a 30% call deflection rate.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Fractal Analytics
Provider: AWS
Maturity: Pilot
Linked source: AWS case study

During the pilot, nearly 500 knowledge workers adopted the solution

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Knowledge management
2Contact center support
3Self-service automation

Fractal Analytics built Knowledge Assist on AWS using Amazon Bedrock to run large language models.
The application uses Amazon EKS for the SaaS application layer, Amazon ECS for connectors, and Amazon OpenSearch Service for vector and semantic search.
The solution ingestes multiple file formats and supports self-service answers and agent assistance within private, encrypted environments.

Technologies

Amazon Bedrock Amazon EKS Amazon ECS Amazon OpenSearch Service

The system handled hundreds of thousands of queries per month across more than 10,000 documents.
The client observed a 10-15% reduction in average data retrieval time and around a 30% call deflection rate.

Architecture

Knowledge Assist runs LLMs on Amazon Bedrock, uses Amazon ECS to build connectors, Amazon OpenSearch Service for semantic/vector search, and Amazon EKS for the application layer. The platform also uses private endpoints and end-to-end encryption, with PII masked before storage in the analytics layer.

Sources & evidence1

Evidence: Medium65/100Evidence strength