Your store
Catalog, policies, blog, PDFs, order data
Use Cases
See all use casesIndustries
View all industriesCapabilities
See all capabilitiesConnect your Shopify store or paste one script. Zipchat reads your product catalog, policies, and help pages. Your AI agent goes live across website chat, WhatsApp, Instagram, Messenger, and email in 20 minutes. No coding required.
7-day free trial · 30-day money-back guarantee
Your store
Catalog, policies, blog, PDFs, order data
Zipchat AI
Knowledge base + retrieval + reasoning
Every channel
Web chat, WhatsApp, Instagram, Messenger, Email
There is no manual training. No scripting. No cold-start period. Zipchat reads your store and starts working from day one. Here is the exact sequence.
Step 1
Shopify stores install from the App Store and enable Zipchat via App Embeds. Five minutes, no developer needed.
Every other platform takes one JavaScript snippet in your site header: WooCommerce, Wix, Webflow, Magento, BigCommerce, PrestaShop, or anything custom. Ten minutes.
Step 2
Zipchat crawls every product page, collection, policy, blog post, and FAQ. Variants, prices, stock status, and the full catalog feed are all indexed. PDFs, custom text, an optional Google Merchant Feed, and a live PostgreSQL connection cover anything else.
You do not write articles. You do not train the AI manually. The knowledge base builds itself in minutes.
Rescans run weekly on Starter and Growth, daily on Pro, Scale, and Enterprise. Update a product or change a policy, and the AI learns the new version on the next sync.
Step 3
Spend five minutes in the AI Training panel. Write a core prompt: brand name, tone, goals (sell, support, recover carts), and topics that are off-limits. Add chat starters, proactive campaigns triggered by exit intent or scroll, and Shopify discount codes generated in conversation.
For actions, add Custom Tools. Connect a shipping provider, CRM, inventory system, or returns portal in plain English. Order tracking, return flows, handoff rules, and ticketing (Zendesk, Gorgias, Intercom, Freshdesk, Salesforce) all sit in the same panel. No coding. No webhooks.
Test Chat mode previews answers before any customer sees them. If a response is wrong, correct it in one click. The AI never repeats the mistake.
Step 4
One knowledge base, one AI, five channels. A customer asking about sizing on your site, delivery on WhatsApp, or a return on Instagram DM gets the same answer from the same source. No duplication. No per-channel training.
The analytics dashboard tracks every conversation that ends in a purchase: chat-to-sale conversion, AOV from chat, total replies, and the questions and objections that come up most. Monthly reports surface what the AI could not answer, so you can close the gaps.
Zipchat does not guess. It does not use generic AI training data to answer your customers. Every answer it gives comes from content it read on your actual store: your product pages, your policies, your blog, your PDFs. If you change a product description, the AI learns the new version on the next rescan.
Every product page, variant, price point, description, and stock status. The AI can answer "Which one comes in size XL?" and "Is the black version in stock?" accurately.
Return policies, shipping timelines, delivery windows, warranty terms, and FAQ pages. The AI handles WISMO questions and policy queries without a human.
Your core prompt defines the AI's tone, personality, and goals. The AI speaks in your brand's voice on every channel, every conversation.
Upload PDFs, paste text, or add individual URLs. Ingredient guides, size charts, assembly manuals, press releases - anything not crawlable goes in here.
Connect a Google Merchant Feed for precise product matching with GTIN and ID fields. Products from the feed are prioritized in AI Search results.
Connect a PostgreSQL database for real-time order history, customer data, and inventory queries. The AI runs read-only queries and answers from live data, never exposing raw SQL or schema.
Zipchat uses RAG (Retrieval Augmented Generation) internally. The AI answers in plain language using the content you provided. Hallucination guards prevent it from inventing information not in your knowledge base. When it cannot find an answer, it says so and escalates to your team.
Each time a shopper sends a message, Zipchat runs a six-stage retrieval loop. The loop retrieves the most relevant catalog page or policy, verifies it against a confidence threshold, then either delivers a grounded answer or escalates to a human - all in real time, with no manual routing required.
The raw message is classified by topic, language, and urgency. Ambiguous phrasing (for example "where is my order") is resolved against conversation history so follow-up questions stay in context.
Semantic vector search and keyword BM25 search run in parallel across your indexed catalog, policies, and blog. The top candidates from both are merged and re-ranked by relevance score.
The top-ranked chunks are assembled into a prompt context alongside the live conversation history. The window is sized so the model always has enough signal without hallucinating peripheral details.
The language model drafts an answer constrained strictly to the assembled context. It formats the reply in the shopper's language and applies your brand tone guidelines.
A confidence score is evaluated against your configured threshold. If the answer meets the threshold, it proceeds to delivery. If not, it forks to the escalation path.
Confident answers are sent to the shopper with a citation reference, in their language and your brand tone. Product links and stock status are included where relevant.
Uncertain answers trigger a handoff to a human agent via Zendesk, Gorgias, Intercom, Freshdesk, or Salesforce, with the full conversation context pre-loaded so the agent never has to ask the shopper to repeat themselves.
You configure the AI once. It then runs natively across every surface where shoppers reach you - the website widget, WhatsApp, Instagram, Messenger, email, and a REST API for custom integrations. Same knowledge base, same brand voice, same accuracy threshold.
Answers questions, recommends products, recovers carts on-site in real time. One line of JavaScript embeds the widget. Brand-matched out of the box.
Reactive support and proactive cart recovery campaigns. Top campaigns recover 20-39% of abandoned carts.
Converts followers into buyers. Handles product questions from DMs and post replies automatically.
Handles customer inquiries from your Facebook Page and ads in real time, with full conversation context preserved across sessions.
The AI reads incoming customer emails and responds using your knowledge base. Multi-turn threads supported. Drafts can be reviewed in Zendesk, Gorgias, or Freshdesk before sending.
Custom integrations - in-app chat, kiosk surfaces, voice transcripts, or your own native app. JSON in, JSON out.
Large language models hallucinate when left unconstrained. Zipchat applies three independent guardrails that together prevent fabricated product, price, or policy answers from reaching your customers.
RAG grounding
Every answer is composed only from retrieved chunks that exist in your knowledge base. The model is instructed not to generate facts that are absent from the retrieved context. No internet browsing. No generic training data.
Confidence threshold
Before delivery, each draft is scored. If retrieval confidence falls below your configured minimum, the AI does not guess - it acknowledges uncertainty and escalates to a human agent. You set the threshold per use case.
Source citation
Every answer surfaces the exact source chunks used to generate it. Customers can see where the information came from. Your team can audit answer quality and identify knowledge gaps without replaying conversations.
What happens when Zipchat does not know the answer?
It says so - in natural language - and routes the conversation to a human agent. The handoff includes the full conversation transcript, the retrieved context, and the confidence score, so the agent has everything needed to resolve the issue immediately.
These are measured outcomes from stores using Zipchat today. Conversion rates are calculated as the share of chat conversations that result in a purchase. Cart recovery rates are from WhatsApp campaigns sent to customers who abandoned a cart.
The industry average ecommerce conversion rate is 2-3%. Top stores on Zipchat reach 25-37% from AI chat conversations. WhatsApp cart recovery at 20-39% outperforms email recovery (3-5% industry average) by 4-8x.
See the full results across 2,000+ stores.
Browse success storiesZipchat runs on Shopify natively. For every other platform, one JavaScript snippet is all that is needed. There is no plugin to maintain, no theme to modify, no developer dependency.
Shopify
1-click install, Built for Shopify certified, checkout support on Plus
WooCommerce
JavaScript snippet, full AI on your WordPress store
Wix
JavaScript snippet via site settings
Webflow
Paste snippet in page settings
Magento / Adobe Commerce
Snippet in header
BigCommerce
Script tag injection
PrestaShop
Header snippet or module
Custom platforms
Any platform that accepts JavaScript
Zipchat connects to Zendesk, Gorgias, Intercom, Freshdesk, and Salesforce for human escalation. It also connects to any external API via Custom Tools. Connect your shipping provider, loyalty program, CRM, or inventory system using plain-English instructions. The AI calls the API during the conversation and shares the result with the customer.
See all integrationsYour customer conversations and store knowledge base are sensitive. Zipchat is architected with security as a first principle - not an upgrade tier.
All stored knowledge base data and conversation logs are encrypted with AES-256. All API traffic uses TLS 1.2+.
Each Zipchat account has its own isolated vector store. Your knowledge base is never shared with, or visible to, other Zipchat customers.
Data processing agreements available. Conversation data can be auto-deleted after 30, 60, or 90 days. EU data residency available on Enterprise plans.
Integrate with your identity provider via SAML or OAuth. Assign view-only, editor, or admin roles to team members without sharing credentials.
Shopify stores go live in approximately five minutes. Install the app, enable the widget via App Embeds, and Zipchat starts crawling your store. On all other platforms (WooCommerce, Wix, custom sites), paste one JavaScript snippet into your site header. The full process takes under ten minutes. The knowledge base builds in the background as the crawl completes.
Zipchat only answers from the content in your knowledge base. It does not invent information. If a customer asks something not covered in your store's pages, policies, or custom content, the AI says it does not have that information and escalates to your team. You can correct any response you disagree with in one click. The AI will not repeat that mistake. Proactive Knowledge Gap Detection also runs automatically: it scans recent conversations, flags gaps, and creates correction suggestions for you to review without reading every chat manually.
When the AI reaches its limit, it collects the customer's email and forwards the full conversation context to your support team. Team members can take over any live conversation directly from the Zipchat dashboard. Zipchat connects to Zendesk, Gorgias, Intercom, Freshdesk, and Salesforce if your team uses a ticketing system. You can also take over manually at any time, regardless of whether the AI triggered the escalation.
Zipchat detects the customer's language automatically and responds in the same language. It supports 95+ languages out of the box with no configuration required. For stores with localized URL structures (for example /fr/ for French, /de/ for German), Zipchat's Localisation feature ensures French customers get answers from French content and German customers get answers from German content, rather than a mixed-language response.
Yes. Depending on your plan, you can create multiple agents. Each agent has its own knowledge base, training, widget, and channels. Growth plan supports five agents. Pro supports ten. Scale supports twenty. This is useful for brands with separate Shopify stores per market, or holding companies running multiple DTC brands from one Zipchat account.
Zipchat scales automatically. There is no traffic cap on conversations. Your AI reply quota (per your plan) is the only limit, and you can increase it at any time by upgrading or enabling auto-charge for additional reply packs. During BFCM, many stores run proactive chat campaigns (exit intent, scroll triggers) and WhatsApp cart recovery sequences simultaneously. The AI handles concurrent conversations without degradation.
The AI can explain your return policy, initiate a return request via a Custom Tool connected to your returns platform (for example Loop Returns or your own returns API), and provide return label URLs. It cannot directly process a refund in your payment gateway. For refunds, it collects the customer's details and escalates to your team with full conversation context, so the agent does not need to ask the customer to repeat themselves.
The AI will say it does not have that information and offer to connect the customer with your team. It will collect the customer's email, summarize the conversation, and forward it via your escalation channel (Zipchat inbox, Zendesk, Gorgias, or email notification). You also receive a notification in your dashboard. The missed question is flagged in the Proactive Knowledge Gap Detection queue, so you can add the answer to your knowledge base and prevent the same gap from occurring in future conversations.