Case study • Ivy City

Ivy City's catalog kept growing; the team didn't.

78.8%

chat containment
18,249

conversations handled across chat, email, and SMS in five months
9,847

FAQs answered

Ivy City Co is a direct-to-consumer fashion brand known for feminine, modest dresses for women and girls, including its signature matching mother-and-daughter styles. Selling primarily through its own Shopify store, the brand has built a loyal community around limited-run drops that release in set quantities and routinely sell out before restocking. That model concentrates demand into short windows, driving spikes in order-status, sizing, restock, and exchange questions from a highly engaged audience that expects fast, personal answers across email, chat, and social.

Shopify
Gorgias
Klaviyo
Redo
Tapcart

Challenge

Ivy City was founded by three women who believed dressing up should still feel magical, even in the middle of motherhood. That vision produced a catalog of modest, feminine dresses in over 20 sizes, including maternity and curve, with new seasonal designs and weekly launches.

The launch cadence and size range were good for the brand. They were hard for the support team. Every new product page needed accurate sizing information in the knowledge base. Every new design sparked a wave of questions: Does it have pockets? Is it true to size? When does it ship? When a customer couldn’t find the answer, they opened a ticket. When the knowledge base had stale or duplicate entries, the agent couldn’t find the answer either.

Returns and exchanges ran on manual workflows built for a smaller operation. Agents handled each case individually, routing tickets through Gorgias by hand. 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. The team was contending with the operational reality of a brand that had grown faster than its support infrastructure.

Solution

Phase 1: Returns, exchanges, and order management

Kodif built automated policies for returns and exchanges, connected to Redo for return eligibility lookups and Shopify for order data. The team went from dry-run testing to live automation.

Returns/Exchanges: Customers navigating a process that previously required agent intervention at every step
Cancel Order: A 4.22-star average CSAT
Update Address: A 4.57-star average CSAT, the highest of any policy, allowing customers to seamlessly and naturally update their address

Phase 2: Sizing and product knowledge

Size Recommendation became the most-executed policy. For a brand with 20+ sizes and new designs every week, keeping that knowledge current was a structural problem. Kodif now ingests Ivy City’s product page data daily, so the AI has current sizing, preorder status, and product attributes without requiring manual knowledge base updates.

Phase 3: Email and SMS coverage

Kodif extended coverage to email and SMS. SMS carried WISMO and cancel order requests with an 83.0% containment rate, the highest of any channel, reflecting a tightly scoped workflow the AI handles consistently. Email containment reached 65.0% across 2,692 eligible conversations.

Lauryn Dawes

Director of Customer Experience

Results

Across the five-month window from January to June 2026, Kodif handled conversations across chat, email, and SMS, executing 3,192 automated policy actions and answering 9,847 FAQ queries without agent involvement.

79%

chat containment

3,192

policy actions executed

9,847

FAQs answered

The policy mix reflects what Ivy City’s customers actually ask about. Sizing was the top topic category followed by. Returns and exchanges, and Promo-related questions. These are predictable, high-volume workflows, and KODIF is handling them all!

Next Up

1

Expanded product attribute coverage as the catalog grows through new seasonal launches
2

Deeper SMS automation for additional order lifecycle events beyond WISMO and cancel
3

Dedicated Gorgias queues for Kodif-synced conversations to improve routing precision
4

CSAT improvement focus on WISMO, where customers expect delivery precision the shipping data doesn't always provide

The takeaway for apparel and DTC brands

Ivy City’s case is a common one: a brand that grew its product line faster than its support infrastructure. The bottleneck wasn’t agent quality. It was volume and repetition. Sizing questions, returns requests, and WISMO queries scale with the catalog; agent headcount doesn’t automatically follow.

What the implementation makes clear is that the workflows worth automating first are the ones with a verifiable outcome. Can the AI look up an order? Check return eligibility? Confirm the right size for a specific dress? Those are automatable. The CSAT data at Ivy City confirms the pattern: the clearer the outcome, the higher the satisfaction.

For CX leaders managing a growing SKU count and a seasonal launch calendar, the question isn’t whether to automate. It’s which workflows are repeatable enough that a human shouldn’t be the one executing them every time.