High-SKU ecommerce
Help shoppers narrow large catalogues by use case, size, material, category, delivery constraints, and product attributes instead of forcing them through filters alone.
AI chat for ecommerce, product catalogues, trade sites, and support-heavy websites
Launch a website assistant you can prove before customers use it.
AI Website Chat Agent helps teams answer product, delivery, compatibility, setup, and support questions from approved website knowledge, files, sitemaps, and supported product feeds — with no-send testing before the widget goes public.
Catalogue-readyuse product feeds and attributes for safer buying guidance
Source-backedanswer from approved knowledge, not broad web guesses
Reviewableweak, risky, or unsupported answers become improvement actions
No credit card needed to start. Paid plans upgrade by card only when approved and configured.
Built for sites where wrong answers cost trust
Explore Product AdvisorFor ecommerce, technical catalogues, trade portals, agencies, and support-heavy websites.
Generic chatbots can sound confident while missing the commercial detail customers actually need. AI Website Chat Agent is positioned for teams that need grounded answers, visible source control, and a safe launch path before automation touches real visitors.
Technical product suppliers
Use approved product data, manuals, FAQs, and policy pages to answer careful specification questions while declining details that are not confirmed by sources.
Agencies and implementers
Offer clients a reviewable AI chat rollout: source import, instruction setup, widget styling, safe test questions, missing-knowledge fixes, and go-live evidence.
Every answer should be traceable before it earns trust.
The platform connects the visitor question, the approved sources used, the evidence state, the missing-knowledge action, and the safe widget test so site owners can improve weak answers before launch.
Example visitor asks
“Which delivery options do you offer for garden furniture, and can I return it if it does not fit?”
Example source-backed answer
“The approved delivery guide says large-item delivery is available on selected products, and the returns policy explains how to report a problem if the item does not fit. I cannot confirm room access or fit from the current sources, so I can pass that to the team before you order.”
Returns policyDelivery FAQProduct feed
Source manifestApproved sources, import state, and excluded/unsafe targets stay visible
TraceableTranscript reviewVisitor question, answer, citations, and follow-up visible to the owner
ReviewedSafe widget testPreview the exact widget response before go-live
No-sendExample improvement action: add a clearer “large-item collection” FAQ before launch.
How the system delivers for real niches
01Ecommerce
Product-aware answers
Use product rows, category context, variants, dimensions, policies, and source pages to answer buying questions more safely than a general FAQ bot.
02Operations
Review before exposure
Run realistic no-send questions, inspect source evidence, approve knowledge, and fix gaps before visitors see the widget.
03Growth
Improve from conversations
Turn unanswered, weak, risky, or off-brand replies into specific knowledge, product-feed, FAQ, instruction, or handoff improvements.
Content resources
View all resourcesLatest resources
The content space is ready; placeholder links point to honest empty-state pages until reviewed posts are published.
Your website and catalogue already hold the answers. The problem is getting them to visitors safely.
Reduce repeat support
Answer recurring delivery, returns, setup, product, and service questions from reviewed sources instead of making the team repeat the same explanations.
Guide complex decisions
For high-SKU and technical stores, help visitors compare options, understand attributes, and know when a human should confirm fit or compatibility.
Launch with control
Preview the widget, inspect source evidence, approve handoff boundaries, and keep public answers constrained to what the business is ready to support.
From website knowledge to working assistant
Add your website
Create your account and connect the site the assistant should support.
Approve knowledge
Import pages, FAQs, files, sitemaps, and supported feeds through a reviewed source workflow.
Shape the widget
Match greeting, colours, prompts, privacy text, and lead capture to your brand.
Test safely
Ask realistic questions in no-send preview before go-live.
Improve continuously
Use conversation review to find missing or weak answers and tune instructions.
Import the knowledge that makes the assistant commercially useful.
The assistant is built around source approval and visible ingestion state. A service website may need pages, FAQs, files, and policy notes; an ecommerce store may also need product-feed rows, attributes, variants, delivery constraints, and category-specific buying guidance.
Grounded knowledge
Use public pages, FAQs, policies, uploaded files, XML sitemaps, manual answers, and supported product feeds as approved sources for customer-facing responses.
Brand and boundary control
Set assistant name, greeting, colours, prompt chips, privacy copy, lead capture, answer style, escalation rules, and when the assistant should decline or hand off.
Operational review loop
Review transcripts, source evidence, leads, unanswered questions, risky wording, stale sources, and missing product fields as a practical improvement queue.
Start free, then choose Website or Commerce capacity based on real usage.
Plans now account for more than chat volume. Website plans suit service, brochure, docs, agency, and support sites. Commerce plans add product-feed capacity, catalogue units, feed-source limits, refresh frequency, and product-answer evidence for ecommerce or high-SKU catalogues.
Free Trial
£0/mo
- One website
- Small approved knowledge set
- Sample product-feed testing only
- No credit card needed
Website Starter
£29/mo
- 1,000 chats/month
- 50 knowledge pages
- Website knowledge, no full product-feed indexing
Website Growth
£79/mo
- 5,000 chats/month
- 250 knowledge pages
- More websites, team users, and review capacity
Commerce Starter
£99/mo
- 3,000 chats/month
- 2,500 product rows
- Product-feed import and catalogue-aware answers
Commerce Growth
£199/mo
- 12,000 chats/month
- 25,000 product rows
- Daily refresh and richer product-answer evidence
Commerce Scale
£399/mo
- 50,000 chats/month
- 100,000 product rows
- Large catalogue, technical, B2B, or managed workflows
Upgrade recommendations use the largest pressure point: chats, websites, knowledge pages, product rows, feed sources, refresh cadence, or catalogue units. No live charge is created until a paid card upgrade is explicitly approved and completed.
Built around approval, boundaries, and visibility.
Approved sources first
The assistant uses the website pages, files, FAQs, sitemaps, manual answers, and supported product feeds the owner has approved. Blocked, stale, or missing sources are surfaced instead of hidden.
Safe fallback behaviour
When a product specification, delivery constraint, price, or policy is not confirmed by approved knowledge, the assistant should say so and offer the next safe action rather than guessing.
Commercial control
Free starts without a card. Paid upgrades, outbound customer sends, and public launch remain controlled by explicit setup and approval gates so teams can prove the experience first.
Questions buyers ask before launch
What makes this different from a generic chatbot?
The product is designed around approved sources, product/catalogue context, safe no-send testing, and owner-visible review states. It is a website/product assistant, not a black-box bot bolted onto the footer.
Can it help with product questions?
Yes, when product data is connected and approved. It can use product rows, attributes, category notes, delivery/returns policies, and setup guidance; if a fact is missing, it should decline safely or route to the team.
Who is it best for?
Ecommerce stores, high-SKU catalogues, technical suppliers, trade portals, agencies, and service businesses with enough website knowledge to support useful answers.
Ask the AI assistant before you commit.
Use our own no-send AI assistant for sales questions, onboarding guidance, help, support, testing, knowledge import, and go-live boundaries without triggering customer messages, live billing, DNS, ads, or deployments.