Normalized claim
Average call handling time: 26% decrease
a 26 percent improvement in average call handling time
Tower Insurance & Deloitte: AI-powered omnichannel contact center with Amazon Connect Customer
Use case typeContact center modernizationUpdated Jul 15, 2026
Tower Insurance, a New Zealand insurer serving about 300,000 customers, modernized its contact center to replace fragmented tools and reduce swivel-chair work for agents. The company implemented a unified AI-powered omnichannel platform with Amazon Connect Customer and Amazon Bedrock, adding call transcription and summarization, near real-time agent assistance, and generative AI evaluation forms. Tower and Deloitte ran the work through a Business Value Realization program and a 10-week AI sprint, then expanded the platform to email and campaign capabilities.
- Organization
- Tower Insurance
- Industry
- Insurance
- Location
- New Zealand
- Published
- July 2026
Reported outcomes
492,000 minutes
customer minutes savedTime & speed
−26%average call handling time−18%average email handling time4-30%automated performance evaluations coverage4 pointsnet promoter score8,200 hoursagent hours freed up
Strategic outcomes
Customer experience & trustImproved service quality and responsivenessScale & capacityExpanded AI use across multiple contact-center workflows
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Average email handling time: 18% decrease
an 18 percent improvement in average email handling time
Normalized claim
Automated performance evaluations coverage: 4-30% increase
coverage of automated performance evaluations rose from 4 percent to 30 percent
Normalized claim
Net promoter score: 4 points increase
NPS rose by 4 points
Normalized claim
Agent hours freed up: 8,200 hours increase
resulted in 492,000 customer minutes saved and 8,200 agent hours freed up
Normalized claim
Customer minutes saved: 492,000 minutes increase
resulted in 492,000 customer minutes saved
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Tower Insurance
- Provider
- AWS
- Maturity
- Exploring
- Linked source
- AWS case study
The company implemented a unified AI-powered omnichannel platform with Amazon Connect Customer and Amazon Bedrock, adding call transcription and summarization, near real-time agent assistance, and generative AI evaluation forms
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 3
- 1Contact center modernization
- 2Customer service automation
- 3Quality management
- Fragmented contact-center tools created swivel-chair activity.
- Call and email handling times were too long.
- Quality assurance covered only a small share of interactions.
- Managers lacked data to coach and staff effectively.
- Built a unified omnichannel agent workspace on Amazon Connect Customer.
- Used Amazon Bedrock for generative AI components.
- Implemented AI-driven post-call transcription and summarization.
- Added near real-time agent assistance through Amazon Connect Customer agents.
- Introduced generative AI evaluation forms to expand QA coverage.
- Extended the solution to email and campaign capabilities.
- Average call handling time improved by 26%.
- Average email handling time improved by 18%.
- Automated performance evaluations increased from 4% to 30%.
- Net Promoter Score increased by 4 points.
- 8,200 agent hours were freed up.
- 492,000 customer minutes were saved.
Architecture
Tower and Deloitte replaced fragmented contact-center systems with a unified omnichannel agent workspace on Amazon Connect Customer. The deployment used Amazon Bedrock for generative AI, AI-driven transcription and summarization, near real-time agent assistance via Amazon Connect Customer agents, and generative AI evaluation forms, then extended into email and campaign capabilities through a 10-week AI sprint and BVR program.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jul 15, 2026Publisher: AWSEvidence: VendorConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.