Product Catalog & AI Quotes Guide

How Velaro handles large product catalogs

Velaro's AI works with product catalogs of any size — from a few dozen items to over a million SKUs. It does this using a two-layer approach: a searchable index for product discovery, and live data queries for pricing and availability. You never need to upload your entire catalog manually.

When Velaro indexes your store (via a connected integration or the web crawler), it reads your product names, descriptions, specifications, categories, and features. This powers semantic search — meaning your customers can ask in plain English ("I need something for outdoor use under $50") and the AI finds the right products even if the customer doesn't know the exact name or SKU.

Why pricing and stock are handled differently

Product descriptions and specs rarely change. Prices and inventory change constantly. Velaro keeps these separate on purpose:

• Product information (names, specs, features, categories) is indexed for fast semantic search
• Pricing and stock are always queried live at the moment of the conversation — never stored as static data

This means a quote from Velaro's AI always reflects your current prices, not a snapshot from last week. It also means you never have to re-sync your catalog when prices change.

What the AI can do with your product data

Once your catalog is indexed, your bot can:

• Search products — find items by description, use case, category, or SKU
• Compare products — give customers a side-by-side comparison of 2–4 items (specs, price, features)
• Create quotes — build a formatted quote with line items, quantities, and totals
• Apply volume pricing — if you configure pricing tiers, the quote respects them automatically
• Save quotes — quotes are saved so your agents can follow up or send them by email

These features work for any business — you do not need BigCommerce, Shopify, or any specific platform to use them.

How to get more accurate results

The AI is only as good as the data it can find. Here is what you can do to improve accuracy:

1. Connect your platform integration

If you use BigCommerce, Shopify, WooCommerce, Magento, Square, or NetSuite, connect the integration. This gives the AI direct API access to your live catalog, real-time stock levels, and customer-specific pricing — far more accurate than web crawling alone.

2. Upload a pricing document

If you have a price list in Excel, CSV, or PDF format, upload it in Admin → Knowledge Base → Documents. The AI will use it as a reference when building quotes. Update the file whenever prices change.

3. Configure a custom pricing API

For advanced setups — especially if you have customer-specific contract pricing, volume tiers, or a separate ERP — you can point Velaro at your own pricing API endpoint. The AI will call it in real time during a conversation. Contact support to set this up.

4. Publish detailed product pages

The web crawler indexes your public product pages. The more detail on each page (specs, dimensions, materials, use cases, FAQs), the better the AI's search and comparison results. Thin pages produce thin answers.

5. Set up PricingConfig rules

In Admin → Quote Settings, you can configure markup percentages, tax rules, minimum order quantities, quote validity periods, and custom header/footer text on quotes. These apply automatically to every quote the AI generates.

What happens if a product isn't in the index yet

If a customer asks about a product the AI hasn't indexed, it will say so rather than guess. It may suggest similar products it does know about. To add a product: publish it on your website (the crawler picks it up within 24 hours), add it to your connected platform integration, or upload a document that includes it.

Accuracy expectations by setup

How the AI determines price during a comparison — priority order

When a visitor asks for pricing as part of a product comparison, the bot works through this sequence in order and stops at the first source that returns a result:

1. Live integration (highest priority) — if your store is connected via Shopify, BigCommerce, WooCommerce, or a custom pricing API, the bot calls it in real time. The price shown is always current. After returning a comparison, the bot will automatically fetch the live price if the visitor asks about a specific product.

2. Indexed catalog snapshot — if no live integration is connected, the bot retrieves pricing from the most recent crawl of your website, your uploaded price lists (PDF, CSV, Excel), or any documents in your knowledge base. The bot will include this disclosure: "The price shown is from our catalog as of our last update — for the most current price please confirm with us directly or visit our website."

3. Handoff — if no price data exists at all (product not indexed, no live integration, no document), the bot will not invent a number. It will say: "Pricing for this product depends on your configuration — let me connect you with a specialist." and trigger your escalation flow.

This order ensures customers are never given stale data without disclosure, and your team never has to correct a bot that invented a price.

The AI will not guess on pricing

If the AI cannot find a definitive price for a product — because it is not in the index, not available via a live API, or requires a custom quote — it will tell the customer that and offer to connect them with a human agent or send a quote request. It will never invent a price.

Large catalog recommendations