01
Order status and delivery tracking
Checks the customer’s email and order number, reads fulfilment and courier tracking from your store, and explains where the parcel is and what happens next.
- Home
- AI Services
- AI Chatbot Development
- AI Chatbot for Ecommerce
AI chatbot for ecommerce
An AI chatbot for ecommerce that knows your orders and your catalogue
Answer “where is my order?”, sizing and returns questions at any hour, and turn product questions into sales rather than tickets. We build a custom AI chatbot on your Shopify, WooCommerce, BigCommerce or Magento store that answers from your own orders, catalogue and policies, and hands anything awkward to your team.
Thanks, that’s with us.
A senior engineer reads every enquiry and will reply as soon as they can.
Want to take it further now?
Book a time that suits you, or send a message on WhatsApp.
- Answers from your own store data: orders, products, stock and policies
- Hands off to Gorgias, Zendesk or your inbox with the order attached
- Built to support UK GDPR requirements, with retention set by you
4.8★
From 944 Fiverr reviews
4.6★
Rated on Clutch
2–4 wks
Typical time from discovery to live chats
100%
Job Success on Upwork
Technologies we build with
- OpenAI GPT
- Anthropic Claude
- Google Gemini
- Product catalogue
- Size guides and policies
- Help centre articles
- pgvector
- Pinecone
- Shopify
- WooCommerce
- BigCommerce
- Magento (Adobe Commerce)
- Gorgias
- Zendesk
- Freshdesk
- Intercom
- HubSpot
- Klaviyo
- WhatsApp Business Platform
- SMS
- Returns apps with APIs
- Courier tracking
- n8n
- Make
Quick answer
What is an AI chatbot for ecommerce?
An AI chatbot for ecommerce is an assistant on your online store, and optionally on WhatsApp, that talks to shoppers in natural language and acts on your store data: looking up an order, explaining your returns policy, recommending products from your catalogue or checking whether a size is in stock. Because it is connected to your platform and helpdesk, it answers with facts about this order and this product, not generic text.
SpiderHunts builds these as a service, not an app you configure yourself. We connect the chatbot to Shopify, WooCommerce, BigCommerce or Magento, load your product data, size guides, delivery and returns policies, connect Gorgias or Zendesk for hand-off and Klaviyo for marketing sign-ups, then test it on realistic customer conversations before it goes live.
UK shoppers often ask about their rights, especially the 14-day period to cancel most online orders. The chatbot answers those questions from your own published returns and cancellation policy, in wording you approve, and passes faulty-goods claims, disputes and anything it is unsure about to a person. It does not give legal advice.
This page is about building one for your store. For a wider look at what chatbots do across ecommerce sales and support, read AI chatbots for ecommerce: sales, support and recovery. If you want the same assistant on WhatsApp for in-store customers too, see our WhatsApp chatbot for retail.
What it handles
What our ecommerce AI chatbot does
Most store support is the same handful of questions. These are the jobs online retailers usually give the chatbot first.
02
Returns, exchanges and cancellations
Explains your returns policy, checks whether the order is inside your window, and starts a return in your returns app where it allows that, or passes it to your team.
03
Product recommendations from your catalogue
Asks what the shopper needs and suggests products from your live catalogue, with links, prices and stock. It recommends only what you actually sell.
04
Size, fit and product questions
Answers from your size charts, product descriptions and care notes, and says when it doesn’t know rather than guess.
05
Stock and back-in-stock questions
Checks stock by size and colour, and offers a back-in-stock alert through Klaviyo or your store when the shopper asks for one.
06
Basket recovery
Answers the questions that stop people checking out (delivery cost, dates, payment options) while they are still on the page, and can follow up a basket by email where the shopper has agreed.
07
Consumer rights questions, from your policy
Explains cancellation, refunds and faulty-goods routes using the wording of your own published policies, and hands disputes to a person.
08
Hand-off to a person
Passes complaints, refund disputes, damaged items and high-value orders to your team in Gorgias, Zendesk or email, with the order and chat attached.
The cost of slow answers
Unanswered questions become tickets, refunds and abandoned baskets
A shopper with an unanswered question either leaves or opens a ticket. Either way, someone on your team ends up answering “where is my order?” again.
Without an ecommerce chatbot
- “Where is my order?” emails fill the helpdesk every morning
- Sizing questions go unanswered until the shopper has left
- Returns questions arrive in five channels with no order number
- Product questions after 6pm wait until the next working day
- Peak season means a backlog and slower replies for everyone
- Staff copy tracking links from one screen to another all day
With an AI chatbot for ecommerce
- Order and delivery questions answered from live store data
- Sizing and stock answered while the shopper is still browsing
- Returns started with the order already identified
- Product questions answered at any hour, from your catalogue
- Your team handles complaints and exceptions, not routine questions
- Every hand-off arrives with the order and chat history attached
See how it would work for you
Book a free 30-minute call. We’ll walk you through how an AI chatbot would handle your order, returns and product questions, and tell you honestly if it won’t help.
Examples
Ecommerce chatbot use cases for UK online shops
Illustrative examples of what we build, not client results.
Size and fit before checkout
Answers sizing questions from the brand’s size charts and fit notes, suggests a size based on what the shopper says, checks that size is in stock and offers a back-in-stock alert if not.
Aim: fewer returns because of the wrong size
Lead times and delivery dates
Reads made-to-order lead times and delivery options from the store and explains when an order will arrive, so the team is not chased by email for dates.
Aim: fewer “when will it arrive?” emails
Product matching from the catalogue
Asks about skin type and concerns, recommends products the store actually sells using ingredients from its product pages, makes no medical claims and passes reactions or allergies to a person.
Aim: shoppers find the right product without a ticket
Order status and returns
Checks the customer’s email and order number, gives the tracking status, checks the return window and opens a Zendesk ticket with the order attached when a return needs approval.
Aim: routine order questions answered without a person
Trade account questions
Answers stock, delivery and account questions for logged-in trade customers, and passes quote requests and credit questions to the sales team.
Aim: trade enquiries routed to the right person first time
Skip, pause or change a delivery
Lets subscribers skip or change a delivery through the subscription app’s API where it allows that, explains cut-off times and passes billing issues to the team.
Aim: subscription changes made without an email thread
Questions that block checkout
Answers delivery date, gift message and gift-wrap questions on the product and basket pages, and offers to email the basket to shoppers who opt in through Klaviyo.
Aim: fewer baskets abandoned over a simple question
Peak season cover
Takes the surge of order and returns questions around Black Friday, Christmas and January sales, with hand-off rules tuned so complaints still reach a person.
Aim: the team stays on exceptions during peak
How we work
How we build your ecommerce AI chatbot
Six steps from first conversation to a chatbot that keeps improving. You test it on your own questions before any customer sees it.
Discovery
We look at your ticket history, platform, apps and policies, and agree what the chatbot answers, what it can change and what always goes to a person.
Week 1Design
We write the conversation flows, customer verification, tone of voice and hand-off rules. You approve them before anything is built.
Week 1Build
We connect your store, helpdesk, returns app and Klaviyo, and set up the data handling and retention you choose.
Weeks 2–3Train
We load your catalogue, size guides, delivery and returns policies, then test it on real past questions, including awkward ones, and fix what it gets wrong.
Weeks 2–3Launch
It goes live on your store, often on selected pages or out of hours first, while your team reviews the conversations closely.
Typically week 4Optimise
We review conversations, fill gaps in its knowledge and add new flows as you learn what shoppers ask. Optional monthly support covers this.
Ongoing
Clear scope · Regular demos · Full handover
Is it right for you?
When an ecommerce chatbot is worth it, and when it isn’t
We would rather tell you now than build something your customers ignore.
Worth it when…
- A large share of tickets are order status, sizing or returns
- Shoppers ask product questions outside your working hours
- Your store, helpdesk and returns app have APIs we can connect
- Your policies and product data are written down and current
- Peak season regularly swamps your support team
Not worth it when…
- You get a handful of customer questions a week
- Most enquiries are bespoke quotes that need a person
- Your product data is incomplete or out of date
- The free chatbot in your helpdesk already covers your questions
- You want it live tomorrow with no testing
Make the right choice
Custom AI chatbot vs chatbot app vs live chat
An honest comparison. Many stores combine them: the AI takes routine questions and people take the rest.
| What matters | Custom AI chatbot | Off-the-shelf chatbot app | Live chat team | Email only |
|---|---|---|---|---|
| Hours covered | Whenever you set, including evenings and weekends | Whenever you set | Staffed hours | Replies in working hours |
| Knows your orders | Yes, connected to your store | Depends on the app’s integrations | Yes, staff look it up | Yes, staff look it up |
| Knows your products and policies | Trained on your own catalogue and policies | Limited set-up options | Yes, with training | Yes |
| Connects to your helpdesk and returns app | Built for your systems | Only the integrations it supports | Usually | Not applicable |
| Peak season | Can take many chats at once | Can take many chats at once | Queues build up | Backlog builds up |
| Who sets it up | We build and maintain it | You configure it yourself | You recruit and train | Your team |
| How you pay | Setup fee, optional monthly support, AI usage | Monthly subscription, often by volume | Salaries or an outsourced team | Staff time |
Pricing
How much does an AI chatbot for ecommerce cost?
Cost depends on your platform, how many systems it connects to and how many order actions it can take, not just answer. You get a written quote with scope and exclusions agreed up front.
Free consultation
Free
30 minutes
We look at your ticket mix and store set-up, tell you honestly what a chatbot could handle and follow up with a written estimate.
Get a quote →Ecommerce chatbot setup
From £5,000
Typically 2–4 weeks
Conversation flows, your catalogue and policies, store, helpdesk and Klaviyo connections, testing and launch.
Get a quote →Monthly support
From £200/month
Optional retainer
Conversation reviews, knowledge updates for new products and policies, new flows and technical monitoring.
Get a quote →AI model usage and any app or helpdesk subscriptions are billed by those providers and estimated in your quote. Prices exclude VAT.
Technology
The technology behind an ecommerce chatbot, in plain English
Four parts work together in every conversation. A large language model, the same kind of AI behind ChatGPT and Claude, understands what the shopper is asking. Retrieval over your own data (often called RAG) means it answers from your catalogue, size guides and policies, not from the internet. APIs connect it to your store, helpdesk and returns app so it can look up an order or start a return. Guardrails check who the customer is before sharing order details and send anything outside its scope to a person.
Understanding: language models
- OpenAI GPT
- Anthropic Claude
- Google Gemini
Your knowledge (RAG)
- Product catalogue
- Size guides and policies
- Help centre articles
- pgvector
- Pinecone
Ecommerce platforms
- Shopify
- WooCommerce
- BigCommerce
- Magento (Adobe Commerce)
Helpdesk and CRM
- Gorgias
- Zendesk
- Freshdesk
- Intercom
- HubSpot
Marketing and messaging
- Klaviyo
- WhatsApp Business Platform
- SMS
Orders and automation
- Returns apps with APIs
- Courier tracking
- n8n
- Make
Who it’s for
Ecommerce chatbots for UK online retailers
Questions differ by what you sell, so the flows, hand-off rules and integrations do too.
Fashion & apparelSizing, fit, exchanges and back-in-stock by size.Furniture & homewareLead times, delivery slots and damaged-item claims.
Beauty & skincareProduct matching and ingredient questions, with no medical claims.
Shopify storesOrders, returns and product questions on Shopify.Shops with a high street presenceClick-and-collect and stock questions on WhatsApp.Ecommerce more widelyAI and software for online retail.FAQ
AI chatbot for ecommerce FAQs
Straight answers to what buyers ask before they sign.
Still have a question?
Ask us directly. A senior engineer reads every question.
Ask about your projectHow much does an AI chatbot for ecommerce cost?
A custom ecommerce AI chatbot from SpiderHunts starts from £5,000 to set up, with optional monthly support from £200. AI model usage is billed separately by the provider and estimated in your quote. The cost depends mainly on your platform, how many systems it connects to and how many order actions it can take.
Does it work with Shopify, WooCommerce, BigCommerce and Magento?
Yes. We connect to each platform through its APIs to read orders, products, variants and stock, and use your helpdesk, such as Gorgias or Zendesk, for hand-off. Where a platform or app limits what can be changed automatically, the chatbot explains the next step and passes the request to your team.
Can it answer UK consumer rights and returns questions?
Yes, from your own policy. It explains your returns and cancellation terms, including the 14-day cancellation period for most online orders, in wording you approve. It does not give legal advice, and it passes faulty-goods claims, disputes and anything outside your policy to a person with the order attached.
Will it recommend products we don’t sell or make things up?
No. It is set up to recommend only from your live catalogue and to answer from information you approve. If a question falls outside that, such as a medical claim or a product you don’t stock, it says it doesn’t know and offers a person rather than guessing. We test this before launch.
Is an ecommerce chatbot UK GDPR compliant?
It can be built to support UK GDPR requirements, but compliance depends on how you use it. Customers are verified before order details are shared, only the data needed is sent to the AI model, retention is set by you, and marketing sign-ups reach Klaviyo only with the shopper’s consent.
How is this different from a chatbot app from an app store?
An app is quicker to install and works within the integrations it supports. A custom chatbot is built around your store, helpdesk, returns process and tone of voice, can connect to systems an app cannot, and is tested on your real questions. If an app already covers what you need, we’ll tell you.
Start here
Book a free ecommerce chatbot consultation
Tell us your platform and what customers ask most. We’ll walk you through how it would work, tell you honestly what a chatbot could take off your team, and give you a realistic range.
- You tell us what you needTwo minutes on the form, or a message on WhatsApp.
- A senior engineer reviews itAnd comes back with questions, a realistic range and an honest view on fit.
- Free 30-minute scoping callWe talk through scope, options and a realistic estimate, with no obligation.
Thanks, that’s with us.
A senior engineer reads every enquiry and will reply as soon as they can.
Want to take it further now?
Book a time that suits you, or send a message on WhatsApp.