What challenges did Orion Health face before AI implementation?

Internal knowledge was fragmented across six separate silos. Staff spent at least 15 minutes per silo each day searching for answers, with some spending over an hour. The fragmented process slowed support resolution, affected SLAs, and reduced productivity.

What was the impact of the AI solution?

The support team is expected to reclaim about 50 staff hours per day. Employees can retrieve information from over 500,000 records in under a minute. Support case resolution became faster and more accurate, improving SLA performance. Orion Health also reported stronger internal governance by helping locate sensitive data, remove duplicates, and clean outdated content. Amazon Bedrock was described as roughly 10x more cost-effective at scale than alternative commercial chatbot platforms.

Rank#1,057of 3,449

Source

Scaled productionEvidence: Medium65/100

Orion Health saves 50 staff hours a day with a generative AI chatbot on Amazon Bedrock

Use case typeStaff assistantUpdated Jun 13, 2026

Orion Health is a New Zealand-based healthcare software company that supports public health systems globally and needed to improve access to internal knowledge spread across six silos. Employees had to search multiple sources for technical documentation, past support cases, and other records, which slowed case resolution and hurt SLA performance. The company built Oribot, an internal generative AI chatbot, to help staff retrieve accurate answers from more than 500,000 internal records quickly and securely.

Organization: Orion Health
Industry: Healthcare
Location: New Zealand
Published: May 2026

Reported outcomes

10x

costCost savings

Strategic outcomes

Speed & agilityFaster support case resolutionBetter decisions & insightImproved answer accuracyRisk & complianceStronger internal governanceCost efficiencyMore cost-effective chatbot platform

Catalog median for cost savings deployments: +53.5% across 16 reported metrics. Compare benchmarks →

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

Normalized claim

Cost: 10 x increase

AWS Solutions Case StudiesMay 11, 2026Case studyInferred claimMedium evidence strength

Amazon Bedrock was described as roughly 10x more cost-effective at scale than alternative commercial chatbot platforms.

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

Customer: Orion Health
Provider: AWS
Maturity: Scaled Production
Linked source: AWS Solutions Case Studies

Amazon Bedrock was described as roughly 10x more cost-effective at scale than alternative commercial chatbot platforms

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Internal knowledge assistant
2Support workflow automation
3Enterprise search

Internal knowledge was fragmented across six separate silos.
Staff spent at least 15 minutes per silo each day searching for answers, with some spending over an hour.
The fragmented process slowed support resolution, affected SLAs, and reduced productivity.

Orion Health worked with AWS Prototyping and Cloud Engineering (PACE) to build Oribot using Amazon Bedrock and retrieval-augmented generation (RAG).
The solution uses a React front end, AWS Lambda for session management, a vector database for semantic search, and runs inside Amazon VPC for controlled data access and compliance.
The chatbot is designed to retrieve answers from over 500,000 internal records and surface context from across support history and product documentation.
AWS helped refine the architecture, prompt tuning, retrieval workflows, and system integration, enabling a working prototype in about two months.

Technologies

Amazon Bedrock AWS Lambda Amazon DynamoDB Amazon VPC RAG vector database React

The support team is expected to reclaim about 50 staff hours per day.
Employees can retrieve information from over 500,000 records in under a minute.
Support case resolution became faster and more accurate, improving SLA performance.
Orion Health also reported stronger internal governance by helping locate sensitive data, remove duplicates, and clean outdated content.
Amazon Bedrock was described as roughly 10x more cost-effective at scale than alternative commercial chatbot platforms.

Architecture

Oribot is an internal generative AI chatbot built with Amazon Bedrock and RAG. It uses a React front end, AWS Lambda, a vector database for semantic search, and runs within Orion Health's Amazon VPC with secure data access policies.

Sources & evidence1

Evidence: Medium65/100Evidence strength