Perry Ellis · Yotpo
1.4% — conversion rate
Independently reportedRetail & e-commerce · Customer support · NLP · Scaled
Customer support
Ivy City · KODIF
Vendor-reportedRetail & e-commerceScaledMid-marketNLPLive 2026
Vendor-reported. The customer is named and the numbers are quoted from the source page, but the account comes from the vendor. No independent confirmation.
Reported by kodif.ai (vendor self-report). Checked against the source page on 2026-09-06. 8 of 8 figures below appear on that page word for word.
The problem
Growing product catalog and launch cadence overwhelmed the support team with inquiries, leading to manual workflows, outdated knowledge, and inefficient handling of returns and sizing questions.
What was deployed
- Vendor
- KODIF
- Products
- Kodif
- Technique
- NLP
- Build or buy
- Bought off the shelf
- Deployment
- SaaS
- Data used
- Ivy City’s product page data daily
- Scale
- 18,249 conversations handled across chat, email, and SMS in five months
What changed
Each row is quoted from the source. Figures we could not find on the page in those words are marked — they are kept, not deleted, so you can judge them.
78.8chat containment
“78.8% chat containment”
Quoted word for word from the source
18,249conversations handledfive months
“18,249 conversations handled across chat, email, and SMS in five months”
Quoted word for word from the source
9,847FAQs answered
“9,847 FAQs answered”
Quoted word for word from the source
4.22CSAT for Cancel Order
“Cancel Order: A 4.22-star average CSAT”
Quoted word for word from the source
4.57CSAT for Update Address
“Update Address: A 4.57-star average CSAT, the highest of any policy”
Quoted word for word from the source
83.0SMS containment rate
“SMS carried WISMO and cancel order requests with an 83.0% containment rate, the highest of any channel”
Quoted word for word from the source
65.0Email containment rate
“Email containment reached 65.0% across 2,692 eligible conversations”
Quoted word for word from the source
3,192automated policy actions executed
“executing 3,192 automated policy actions”
Quoted word for word from the source
Difficulties and limits
- The Gorgias chat widget had its own complications: it would occasionally show agents as unavailable when they weren’t, and block customer responses mid-conversation.
- keeping that knowledge current was a structural problem
- Every new product page needed accurate sizing information in the knowledge base, and when the knowledge base had stale or duplicate entries, agents couldn't find the answer either.
- Returns and exchanges ran on manual workflows built for a smaller operation, with agents handling each case individually and routing tickets through Gorgias by hand.
- shipping data doesn't always provide delivery precision
Sources
- kodif.ai — vendor self-report, 2026-07-13
Similar deployments
Innovative Dining Group · Revmo AI
1200+ — Reservations booked by AI
Vendor-reportedTravel & hospitality · Customer support · NLP · In production · 2025
Houston Methodist · ActiumHealth
18000000 — staffing capacity created
Vendor-reportedHealthcare · Customer support · LLM · In production
Luxury Bath NJPA · Hatch AI
52% — Overall set rate
Vendor-reportedConstruction & real estate · Customer support · AI agents · Scaled
Radiant Credit Union · interface.ai
60% — Asset growth absorbed without contact center team expansion
Vendor-reportedBanking & finance · Customer support · LLM · Scaled · 2022
Talkmobile · LivePerson
92% — Customer Satisfaction Score (CSAT)
Vendor-reportedTelecom · Customer support · NLP · Scaled · 2014