Customer Experience

Support operations for an international apparel brand

Rebuilt customer support for an international apparel brand, cutting average resolution time from 11 days to two days and helping raise its Trustpilot rating from 3.4 to 4.4 stars. AI drafted responses for 70% of inbound inquiries.

Inquiries handled with AI drafts70%

Average resolution time11 days → 2 days

Trustpilot rating3.4 → 4.4 stars

My role

I audited the customer journey, built the support playbook and triage process, implemented AI drafting and review follow-ups, and recruited and trained an offshore CX specialist to run daily support.

Setup began around June 2025. Ticket resolution comparison: January–June 2025 versus July–October 2025. The specialist now runs daily support, with occasional help from me.

The challenge

An apparel brand selling internationally was handling customer support ad hoc across the team. Replies were inconsistent, issues stayed open, and creative staff were pulled away from product design and launches.

The goal was to give support a clear owner and a repeatable process, then use AI to help that person respond faster.

My approach

Build the playbook first

I audited the customer journey and documented SOPs, response templates, and a tone-of-voice guide. The resources covered common questions about orders, exchanges, returns, and refunds. Edge cases went to a Slack channel for team input.

Give support a clear owner

I recruited and trained an offshore CX specialist for 10 hours per week. They handled 95% of customer interactions using the playbook. Gmail tags made open and resolved tickets visible, and a 48-hour response target set a clear expectation for the team. The workflow is now self-sufficient day to day; I still help occasionally.

Use AI to speed up the first draft

I implemented a tool that drafted responses as emails arrived, covering 70% of inbound inquiries. The specialist could review and send those drafts rather than start every reply from scratch. Automated acknowledgments set expectations while customers waited.

Follow through after resolution

I set up review follow-ups and reached out to past reviewers with unresolved issues. This extended the work beyond closing tickets to repairing customer relationships.

Outcome

Response time: 41 hours → 21 hours.
Resolution time: 259 hours → 41 hours, an 84% reduction.
Ticket resolution: 36% (101 of 284 tickets) → 100% (126 of 126 tickets).
Trustpilot rating: 3.4 → 4.4 stars.
Customer satisfaction: 81% CSAT.

The resolution-rate comparison covers January–June 2025 (before) and July–October 2025 (after). These results reflect the combined changes to staffing, processes, and AI drafting.

What I learned

A clear process makes AI more useful. The playbook, ownership, and triage system gave AI drafting a useful role in the workflow. Faster drafts supported a consistent service process.

A small team can work well with clear boundaries. Documenting routine inquiries and giving exceptions a clear escalation path let one part-time specialist handle most customer interactions while the creative team focused on its core work.

Resolution is part of the brand experience. Following up on outstanding issues created opportunities to rebuild trust. The improvement in the review rating made that work visible beyond the inbox.

END OF LINER NOTES ↟