← All case studies

AI that asks shoppers why they did not buy

Hesient builds AI products for online stores. Sell&Tell serves Shopify merchants, and GreenLync is the same engine built for cannabis retail. Each adds an assistant to the store that asks hesitating shoppers what is holding them back, then turns the answers into a weekly brief of what to fix.

Ongoing engagement with Hesient Inc.

ClientHesient Inc.
DisciplineSoftware development and AI
Our roleProduct engineering: architecture, development and DevOps
PeriodNov 2025 to present
2SaaS products in production
2AI agents on GreenLync
4Live websites and apps
01 / The challenge

Store analytics show what shoppers did, not why they left. Merchants guess, and the usual fix is a discount. Hesient wanted a product that asks the shopper at the moment of hesitation, tells the merchant what to change, and then checks whether the change worked.

02 / What we built
  • Sell&Tell for Shopify: an on-page assistant that notices hesitation and asks why, and a merchant app with a weekly Monday Brief of what to fix and whether last week's change helped. Billed through Shopify.
  • GreenLync for cannabis retail: the same engine, triggered by behavioural signals rather than a timer, with Talk to Your Data analytics and Stripe billing.
  • Two Gemini-backed agents for GreenLync: Lync, an analytics assistant for dispensary operators, and Tokey, a shopper-facing budtender behind a 21+ gate that can read only customer-facing product fields.
  • Both marketing websites, including the Sell&Tell affiliate program with a working application flow.
  • Infrastructure: Cloud Run services and jobs, Cloud SQL PostgreSQL, Redis, and Cloud Build CI/CD with preview deployments.
03 / Stack
Next.jsReactTypeScriptPrismaPostgreSQLRedisNextAuthGeminiClaudeStripeShopifyCloud RunCloud SQLCloud Build
Next project

Have a system to build or fix?

Software, electronics, or both. Your message goes straight to the architect, not a sales team.

Start a project