AI that answers product questions from your catalogue
Answer product questions from your own catalogue, explain saved prices and policies, compare listed items, and hand uncertain questions to your team.
Give shoppers useful answers about the products you actually sell. OctaDezx uses your saved catalogue and business knowledge to explain an item, discuss alternatives and continue the conversation when someone is ready to order.
The everyday problem
A shopper asks whether a boot fits wide feet, whether the black version costs more, or which bag suits a laptop. A generic answer is easy to produce. A correct answer needs the facts your store has supplied.
How OctaDezx helps
Product names, descriptions, saved prices and your policies give the assistant a business-specific starting point. A missing specification should become a clarification or a human handoff, rather than a convincing guess.
How the catalogue becomes a useful answer
- Import and review your products. Connect Shopify or WooCommerce, or maintain products in the dashboard. Imports create local product records. Review descriptions and active status before relying on them in customer conversations.
- Explain documented differences. The assistant can compare the descriptions and saved details of two listed products. Materials, dimensions and fit advice must be present in the catalogue or knowledge base to be used reliably.
- Keep prices and stock in context. Answers use the saved catalogue snapshot. Shopify imports retain variant metadata, but the main product price comes from the first variant. Imported stock can be aggregated and unknown stock is not the same as zero.
- Add the policies a product row cannot hold. Save delivery areas, return conditions and sizing guidance as business knowledge. Keep those policies consistent with the store so a helpful product answer does not become an unsupported delivery promise.
- Continue across connected channels. Use the same business information in website chat, WhatsApp, Instagram DMs and Facebook Messenger after connecting those channels. Channel permissions and delivery rules still apply.
- Escalate missing information. Tell the assistant how to hand over custom requests, undocumented specifications and uncertain availability. Your team can take over with the conversation attached.
From catalogue fact to customer answer
Illustrative example, not a customer transcript. The sample product and its specifications are invented to explain the workflow.
- 1. Save the source. A sample City Backpack lists a 15-inch laptop compartment, water-resistant fabric and a saved price. The catalogue does not claim it is waterproof.
- 2. Answer the question. For "Will it fit my 15-inch laptop?", the assistant can explain the documented compartment. It should ask for measurements if the shopper needs a tighter fit check.
- 3. Compare listed alternatives. If a second listed bag has a documented larger compartment, compare those facts. Do not infer that a more expensive item must be more durable.
- 4. Hand over uncertainty. For "Can I submerge it in water?", explain that water resistance is the only documented claim and route any unresolved request to the team.
A catalogue is a source, not a guarantee
- Imports and supported store events update local data. This is not a fresh store API check on every question, and failed syncs can leave a stale snapshot.
- A variant in metadata does not guarantee variant-level checkout, reserved inventory or a distinct variant price at order capture. Review those details with the customer and your team.
- Do not promise fit, safety, compatibility or availability when the supplied product information cannot establish it. Remove or deactivate discontinued items and check the next conversation.
Frequently asked questions
- How does the catalogue stay current? Use the store connection to import products and review sync results. Supported webhooks can update local records, but availability depends on successful setup and delivery. Recheck important changes in the dashboard.
- Can the AI compare two products? It can compare supplied descriptions and saved facts for listed products. It should not invent a specification, review or performance test to fill a gap.
- What if a product is discontinued? Update or deactivate it in the catalogue and verify the change has reached the local records. Add instructions for substitutes, refunds or existing orders where needed.
- Can it recommend a similar item? Yes, recommendations can draw from your listed products. They depend on the information available and are not a promise that an alternative has identical fit or performance.
- Does this work outside ecommerce? The same approach can explain listed services, menus and offerings. Give it the business facts and decision boundaries relevant to your workflow.
Explore the connected tools
- Ecommerce customer care
- Shopify connection
- WooCommerce connection
- Find products from a photo
- Catalogue training guide
- Plans and pricing
Read more on this
- How to train an AI on your product catalogue
A practical guide to importing product data, writing clear policies, reviewing wrong answers and fixing the five failure modes of catalogue-trained support. - How to reduce Shopify support tickets
Find the sources of Shopify support tickets, remove repeated questions at the source and keep complaints, disputes and uncertain answers with a person.