Catalog-grounded answers
Recommendations are built from your live product data, including variant availability, so the assistant does not push an item that is out of stock.
A chat assistant that knows your catalog. It suggests products, answers size and specification questions, checks order status and passes anything it cannot handle to a human agent.
ShopGuide AI adds a chat widget to your storefront and grounds its answers in your own product data. It reads titles, descriptions, variants, prices, tags and stock levels from Shopify, so recommendations point at products you actually sell and can actually ship. It can walk a shopper through sizing, compare two items, look up an order by number and email, and hand the conversation to a person when the question is a refund dispute or anything else needing judgement.
Recommendations are built from your live product data, including variant availability, so the assistant does not push an item that is out of stock.
Upload size charts, fabric notes, wattage tables or compatibility lists and the assistant answers from them instead of guessing at a plausible number.
A shopper gives an order number and the email on file, and the assistant returns fulfilment status and the tracking link without opening a support ticket.
Escalation rules pass the thread to your team with full transcript context, either into your helpdesk or straight to a WhatsApp conversation.
The assistant replies in the language the shopper used, covering Urdu, English and Roman Urdu, with Arabic available for GCC storefronts.
See what people ask most, which questions end without a purchase, and where the assistant escalated. That list is a product page improvement backlog.
Install the app and let it index your catalog, collections, policy pages and any size or spec documents you upload.
Set the assistant tone, greeting, languages and the rules that decide when a human takes over.
Add the widget to your theme through the app embed and choose which page templates it appears on.
Review transcripts weekly, correct wrong answers, and add missing information to your product content.
A large share of browsing happens late at night. Questions answered at that moment convert; questions parked until morning usually do not.
Sizing, availability and delivery timing account for most inbound chat. Handling them automatically frees your team for orders that need real intervention.
An assistant that understands the catalog can suggest the matching accessory or the next size up at the moment the shopper is deciding.
Guided sizing before checkout removes a chunk of exchange traffic, which matters more in Pakistan where reverse logistics are slow and costly.
[PRICE] per month. A monthly conversation allowance, catalog indexing and English replies for a single storefront.
[PRICE] per month. Higher conversation allowance, multilingual replies, order lookup, human handoff and analytics.
[PRICE] per month. Multi-store and multi-language deployments, custom knowledge sources and priority support.
Answers are grounded in the catalog and documents you provide, and the assistant is instructed to say it does not know rather than fill a gap. The practical safeguard is your own content: if a spec is missing from the product page, add it, and review transcripts for questions that keep returning unanswered.
The widget loads asynchronously through a Shopify theme app embed, so it does not block page rendering. You can also restrict it to specific templates, for example product and collection pages only, to keep it off the checkout path entirely.
Yes. A large share of Pakistani shoppers type Urdu in Latin script, and the assistant reads it and can reply in the same style. You choose whether replies come back in Roman Urdu, Urdu script or English.
Products, variants, inventory, collections, policy pages and, for order lookup, orders matched by number and email. Access is granted through the Shopify app scopes you approve at install, and you can decline order lookup if you would rather keep that behind a human.
We will index your catalog, tune the escalation rules and review the first month of transcripts with you.