Macha

How to Add an AI Chat Agent to Gorgias Chat

Abbas, Customer Support & AI, Macha

Written by

Ankeet Guha, Co-founder & CTO, Macha

Reviewed by

Published July 23, 2026

Updated July 23, 2026

The Gorgias chat widget on your storefront is where a lot of revenue is quietly won or lost. A shopper hovering over checkout asks whether shipping is free, or how long delivery takes to their state, and the speed of that answer decides whether they buy or bounce. An AI chat agent turns that widget from a form that collects questions into a system that actually answers them — resolving the routine ones on its own and handing the tricky ones to a human before frustration sets in. This guide walks through adding a resolving AI agent to Gorgias Chat step by step, using real ecommerce examples like WISMO, returns, and subscriptions, and it stays honest about where the native tools shine and where a dedicated agent layer has the edge.

How to Add an AI Chat Agent to Gorgias Chat

What "an AI chat agent" means on Gorgias

There are two very different things people mean by an AI chat agent, and the difference matters for your CX. A deflection bot shows a shopper some help-center articles and hopes they read the right one — it reduces tickets by getting in the way. A resolving agent reads the shopper's actual question, pulls the relevant fact from your knowledge base or their order, and gives a specific answer that closes the conversation. Gorgias's native AI Agent is firmly in the second camp: per the official chat documentation, it resolves conversations directly on the widget, adapts its tone to be "more conversational, quick and concise" on chat than on email, and hands off to a human when it should.

That distinction — deflection versus resolution versus full automation — runs through everything below, and it's the same lens we use in our Gorgias AI Agent explained walkthrough.

Before you start: the prerequisites

Gorgias's AI Agent on Chat has a few hard requirements worth checking before you click anything. Per the docs, you need:

  • An active AI Agent subscription (it's an add-on to your Helpdesk plan, priced separately).
  • Lead, Admin, or Account Owner permissions.
  • A Shopify store connected to Gorgias. As of this writing the native AI Agent on Chat is Shopify-only — BigCommerce, Magento, and WooCommerce aren't supported for this specific deployment.
  • Chat installed on your website or Help Center.
  • AI Agent onboarded with knowledge — your Help Center articles, website content, or Custom Guidance documents.

That last point is the one that determines whether your agent resolves or flails. An AI chat agent is only as good as what it's grounded in, so populate your Help Center with real answers to shipping, returns, and subscription questions before you flip the switch.

Step 1: Enable AI Agent on Chat

Once the prerequisites are met, turning it on is quick. Per Gorgias's chat setup guide:

  1. Open the dropdown menu in the top-left and select AI Agent.
  2. Under the Deploy section, select Chat.
  3. Open the channel dropdown and choose one or more Chat channels you want the agent to cover.
  4. Toggle Enable AI Agent on Chat to ON.
  5. Select Save Changes.

One thing to know: Article Recommendations are automatically disabled when you enable AI Agent on Chat, so the widget doesn't send conflicting messages. That's the platform gently pushing you from a recommendation bot toward a resolving agent.

Step 2: Ground it in your knowledge (and give it order data)

A resolving agent needs two kinds of knowledge: general policy and shopper-specific facts. Gorgias's AI Agent draws general answers from your Help Center content, website resources, and Custom Guidance documents, and — critically for ecommerce — it can read a shopper's order history after authentication. That's what lets it answer a genuine WISMO ("where is my order?") question with a real tracking status instead of a canned "check your email" reply.

Here's what that looks like in practice with a dedicated agent layer resolving a live shipping question end to end, grounded in the knowledge base with a single tool call:

A Macha chat agent resolving a live conversation: asked about free shipping and US delivery times, it grounds its answer in the knowledge base (one tool call) and returns specific shipping rates and delivery windows.
A Macha chat agent resolving a live conversation: asked about free shipping and US delivery times, it grounds its answer in the knowledge base (one tool call) and returns specific shipping rates and delivery windows.

The pattern to copy: the agent doesn't recite your whole shipping policy — it retrieves the one fact the shopper asked for (free-shipping threshold, US delivery window) and returns it conversationally. That's resolution, not deflection.

Step 3: Set your handoff rules so humans catch the right conversations

The single most important design decision in an AI chat agent isn't what it answers — it's when it stops and hands to a human. Gorgias's AI Agent hands over automatically when it:

  • Lacks confidence in an answer.
  • Hits a listed handover topic you've defined (e.g. chargebacks, wholesale, legal).
  • Can't find relevant knowledge to answer from.
  • Detects the shopper explicitly asking for a human.
  • Identifies anger or frustration in the message.

For an ecommerce store, a sane starting set of handover topics looks like this:

Shopper intentAI resolves?Handoff rule
WISMO / order statusYes — reads order + trackingOnly if tracking is missing/stuck
Free-shipping / delivery-time questionYes — KB-groundedRarely
Return / exchange eligibilityYes for policy; hand off for exceptionsOff-policy or damaged item
Subscription: skip / pause / swapYes if connected to the actionCancellation-save or billing dispute
Chargeback / fraud / legalNoAlways hand off immediately
Angry or escalating toneNoHand off, tag as priority

Handed-over conversations are the release valve that keeps AI from damaging your CX — get these rules right and the agent only closes what it can genuinely close.

The honest part: what native Gorgias AI does well — and the billing catch

Credit where it's due. Gorgias's native AI Agent is a genuinely strong ecommerce agent: it's purpose-built for Shopify stores, it reads order history, it takes actions across 100+ ecommerce tools, and its handoff logic is sensible out of the box. For a lot of stores it's the fastest path from "chat form" to "chat that answers." Our complete guide to Gorgias Automate and AI covers the full native feature set.

Where it gets worth thinking about is how it's billed. Gorgias's AI Agent is priced per resolved conversation — per the official AI Agent pricing page, that's $0.90 on most plans and $1.00 on Starter, with plan tiers running from 90 to 2,500+ interactions per month. A billable resolution happens "when the AI resolves a customer conversation entirely on its own." Conversations it hands to a human are not charged as automated interactions — they simply become standard helpdesk tickets. Overages run $1.00 to $2.00 per interaction on monthly plans (or $0.83 to $1.67 annual), and the Shopping Assistant add-on is $1.50 per interaction across paid tiers.

Per-resolution pricing is clean and easy to reason about, and for pure deflection it's fair. But it ties your bill to a single outcome — the "fully resolved" conversation — and that shapes behaviour. Two things fall through the gap: work that isn't a clean full resolution (a partial answer, a triage, a summary, an internal draft that a human finishes), and multi-step automation where the value is in the actions, not a closed chat. Here's the contrast:

Per-resolution billingPer-action billing
You pay forA conversation the AI closes end to endEach AI action the agent takes
Great atClean deflection / full resolutionDeflection, resolution, and partial + multi-step automation
Partial help (triage, draft, summary)Not the unit of valueCounted as the actions it is
Multi-step workflowsHard to price cleanlyEach step is a metered action
PredictabilitySimple, but outcome-coupledMetered by work done

This is the philosophical difference behind Macha's pricing: credits are consumed per AI action, not per resolution. That frames the three modes distinctly — deflection (steer to an answer), resolution (close the conversation), and automation (take actions across your stack) — rather than collapsing them into one "resolution" line item. Neither model is universally cheaper; they optimise for different work. For a fuller side-by-side of the native numbers, see Gorgias pricing explained.

Where a dedicated agent layer has the edge

Macha is not a Gorgias replacement — it's an AI agent layer that runs on top of the Gorgias you already use, via a live connector. You keep Gorgias as your help desk, your inbox, and your chat widget; Macha reads and writes the same conversations and adds the reasoning-heavy work. Three places that layer earns its keep on chat:

  • Grounding you control. Point the agent at your knowledge base and it retrieves the exact fact a shopper asked for (the screenshot above is a single KB tool call resolving a shipping question), instead of guessing.
  • Custom actions beyond the catalogue. When a WISMO answer needs a lookup in a 3PL or an OMS that isn't a native Gorgias integration, a custom tool turns that REST API into something the agent can call — so "where's my order?" gets a real status, not a handoff.
  • Metered by work, not outcome. Because credits are per action, a triage, a draft-for-a-human, or a two-step subscription change all count as the work they are — useful for stores where a lot of chat value is partial help, not clean full resolutions.

If you're weighing the native agent against a layer, our roundup of the best AI agents and chatbots for Gorgias compares the options, and the Macha–Gorgias integration page shows exactly how the connector wires in. The clean division of labour: let Gorgias own the widget and the inbox; layer an agent on top for grounded resolution, custom lookups, and the automation the native billing model doesn't naturally reward.

FAQ

How do I add an AI chat agent to Gorgias Chat? With an active AI Agent subscription and a connected Shopify store, open the AI Agent dropdown, go to Deploy → Chat, choose your chat channel(s), toggle Enable AI Agent on Chat on, and save. Onboard it with Help Center, website, or Guidance knowledge first so it has something to resolve from.

Does the Gorgias AI chat agent actually resolve questions or just deflect? It resolves. Unlike an article-recommendation bot, Gorgias's AI Agent reads the shopper's question, grounds an answer in your knowledge base, and — after authentication — can read order history to answer WISMO questions with real tracking status. Article Recommendations are automatically disabled when you turn AI Agent on Chat on.

When does the AI chat agent hand off to a human? Automatically when it lacks confidence, hits a handover topic you've listed (like chargebacks or legal), can't find relevant knowledge, detects an explicit request for a human, or detects anger or frustration. Handed-over conversations become standard helpdesk tickets and aren't billed as automated resolutions.

How much does the Gorgias AI chat agent cost? It's priced per resolved conversation — $0.90 on most plans and $1.00 on Starter — as an add-on to your Helpdesk plan, with tiers from 90 to 2,500+ interactions per month and overages of $1.00–$2.00 (monthly) per interaction. Confirm current numbers on Gorgias's pricing page, as tiers change.

Can I add an AI chat agent to Gorgias without replacing Gorgias? Yes. An agent layer like Macha connects to Gorgias as a live connector and runs on top of your existing help desk and chat widget — it doesn't replace them. It adds KB-grounded resolution, custom lookups, and metered-per-action automation, while Gorgias stays your system of record.

Ready to turn your chat widget into an agent that actually resolves? Start a free trial of Macha and connect it to your Gorgias in minutes.

Macha

About Macha

Macha is an AI agent platform that works on top of the help desk you already use — Zendesk, Freshdesk, Gorgias, or Front — and connects to the rest of your stack, even your own internal systems. Its AI agents resolve tickets and automate entire workflows end to end, all set up in plain English, no code. Learn more about Macha →

Zendesk
5.0 on Zendesk Marketplace

Loved by support teams worldwide

See what support teams are saying about Macha AI.

The application seems excellent to me! We are still testing, and we need support for some details and they were extremely efficient too!

Daniela Costa

Daniela Costa

Head of Support, Seabra

Macha has been a great addition to our support toolkit. It generates clear, well-organized responses that fit naturally into our workflow. One feature we particularly appreciate is its ability to automatically reply in the same language as the ticket.

Marius F

Marius F

Support Head, Zentana

We've been using Macha for a little while now and it's been really great addition so far! It's powerful, convenient, and makes getting work done a lot easier for our agents.

Alexander Wedén

Alexander Wedén

Head of Support

Support team is very helpful and responsive. Really enjoy how lightweight this is within Zendesk itself vs other more intrusive tools.

Cathleen Wright

Cathleen Wright

Zendesk Admin, Cortex IO

So far it's pretty good! Our queries are a little nuanced, so we can't always use it, but it's got enough utility for us. It can even incorporate our bilingual country with greetings in a second language.

Jae Oliver

Jae Oliver

Head of Support, Wise

Really enjoying using Macha, it has made a noticeable difference to our support team in a short amount of time. I really like the ticket summary feature, saves us a lot of time.

Harry Jackson

Harry Jackson

Head of Support, Crumb

Macha AI is a great addition to my workspace! It's powerful, convenient, and it really makes productivity so much easier for our agents!

Dave G

Dave G

Head of Support, Cyber Power Systems

Very impressed! AI integration for Zendesk has certainly come a long way and Macha seems to set the standard for now. This will for sure save lot of time in our support team.

Pauli Juel

Pauli Juel

Head of CS, Dokument24

Macha has been working great for us so far! The auto-responses are accurate and our resolution time has dropped significantly.

Lana T

Lana T

Zendesk Admin, Swotzy

Macha AI is a great addition. The knowledge base feature means our agents always have the right answers at their fingertips.

Mischa Wolf

Mischa Wolf

Head of Support, Topi

We're enjoying this integration so far. It's made our support team more efficient and our customers get faster responses.

Paula G

Paula G

Head of Customer Support, Xly Studio

The team enjoys using it. It saves considerable time on common questions and the integration options are excellent.

Kilian Leister

Kilian Leister

Support Head, Didriksons

Ready to supercharge your team with AI?

Get started in minutes. Connect your tools, configure your agents, and let AI handle the rest.

500 free credits · no time limit, no credit card