NewNew: The enterprise guide to Agentic AI — 24 min read.

Read →
Case study · Retail · CCaaS migration

90-day cutover from legacy IVR to Amazon Connect at a national retailer

A national specialty retailer had 90 days to get off a decade-old Avaya IVR before their peak holiday window — and wanted agent assist live on day one. pronix.ai ran the cutover on Amazon Connect with Amazon Q for agent assist, held CSAT flat through peak, and cut AHT by 28%.

Client
National specialty retailer, 2,400 agents
Industry
Retail & Ecommerce
Platform
Amazon Connect · Amazon Q · Salesforce Service Cloud
By pronix.ai CX Engineering8 min readQ4 2025
90 days
Legacy IVR to live on Amazon Connect
-28%
AHT
+1.4pt
CSAT through peak
0
Missed peak days

*Representative outcome; results vary by client, scope and platform configuration.

The challenge

A decade of Avaya customization, 2,400 agents across three sites, and a 90-day window before holiday peak. Prior migration attempts had stalled on IVR parity and integration debt. The stakeholder bar: zero CSAT dip, no missed peak day.

Our approach

Step 01

Cut scope, not quality

We shipped the top 80% of IVR flows in phase one and put the tail in a parallel workstream. That call — made in week two — is what made 90 days possible.

Step 02

Agent assist on day one

Amazon Q for real-time suggestions, tuned on 6 months of anonymized transcripts. Went live with the platform, not six months after.

Step 03

Warm-cutover, not big-bang

Site-by-site cutover on a Sunday-Monday rhythm. Rollback scripts rehearsed twice per site. War room staffed for 72 hours after each site went live.

Step 04

Real-time CSAT signal

Post-call sentiment plus first-hour supervisor sampling. We caught two intent-routing regressions in the first 24 hours and fixed them before they hit peak.

This was the calmest peak we've had in six years — and we changed the platform on the way in.

VP Contact Center Operations
For VP Contact CenterFor CX Platform OwnerFor CIO

Illustrative case study. Scenarios, metrics, quotes and client details are representative composites based on Pronix engagements and industry benchmarks unless a named client is shown with written consent. Outcomes vary by client, scope, data quality and platform configuration. Nothing on this page is a guarantee, warranty or professional advice. See our Terms of Use for the full disclaimer.

Trademarks.

AWS, Salesforce, Microsoft, NICE, Genesys, Kore.ai, Google, ServiceNow, Amazon Connect and logos referenced on this site are trademarks of their respective owners. References are for descriptive purposes only and do not imply endorsement, sponsorship or partnership beyond stated partner relationships. See our Disclosures.

Bring this to your program

Talk to the team that shipped it.

Book a working session with a pronix.ai lead on this practice — we'll map the approach to your platform, industry and constraints.