Case study · E-commerce Brand

$120K saved annually by automating customer support

A fast-growing DTC apparel brand was buried under 12,000+ support tickets a month across email, chat and social DMs. Thinkly built a Shopify-connected AI chatbot and an internal escalation workflow that now resolves 80% of tickets without a human — saving $120K a year in outsourced support fees.

E-commerce fulfilment warehouse with packages on shelves
90%
Faster response time
4,000+
Hours saved per year
$120K
Annual support-cost saving
80%
Tickets resolved without human

The problem

As the brand scaled past $10M ARR, its outsourced support agency became the largest line item after inventory — burning ~$180K/year while median first-response time crept above 14 hours.

80% of tickets were the same five questions: where is my order, how do I return, what size should I buy, do you ship to X, is Y back in stock.

Agents were manually copy-pasting tracking numbers from Shopify into Gorgias and reshipping labels one at a time.

What we built

Thinkly built a custom AI chatbot trained on the brand's product catalog, size guide, shipping matrix and returns policy, deployed on the website, Instagram, WhatsApp and email — with a single shared context so a customer can start on IG and finish on email seamlessly.

The chatbot is wired directly to Shopify, ShipStation and Loop Returns via server-side workflows: order lookups, tracking updates, return-label generation and exchange creation happen inside the conversation, not in a separate portal.

A confidence-based escalation model routes any query the AI is <90% sure about to a human agent with full conversation context pre-loaded — so humans only see the hard 20%.

A weekly quality-review dashboard surfaces the top 20 lowest-confidence conversations so the team can retrain prompts and knowledge base without engineering support.

The results

Services delivered

Our AI chatbot handles 80% of customer queries instantly. Returns, orders, tracking — all automated.
Usman Malik, CEO
Results disclaimer. Case-study figures describe outcomes for this specific client under specific conditions. Your results depend on your data, team, market and execution and may differ materially. See our about page and editorial policy for details.

Book a free consultation →See more case studies