The Commerce Context Engine
The Commerce Context Engine is the data layer every Webscale AI product runs on. It connects your storefront, back-office systems and expert knowledge, normalizes them into one model-ready context and keeps that context current.
What is the Commerce Context Engine?
Section titled “What is the Commerce Context Engine?”The Commerce Context Engine is the Webscale AI data layer for agentic commerce. Every Webscale AI product runs on it, including the AI Shopping Assistant and Audience Segmentation. It connects the systems that hold your commerce data, normalizes that data into one model-ready context and keeps the context current.
Model-ready data is structured, validated and complete enough for an AI agent to read and trust without guessing. The Commerce Context Engine produces it from the data you already have.
What data does the Commerce Context Engine use?
Section titled “What data does the Commerce Context Engine use?”The Commerce Context Engine reads from three kinds of source. Together they hold the facts a shopper’s question depends on.
| Source | What it includes |
|---|---|
| Your store | Product catalog and site content, such as policy and FAQ pages |
| Your back office | Inventory and pricing, plus orders and customer data on Magento and Adobe Commerce |
| Your people | Expert knowledge, support knowledge and documents such as PDFs and spec sheets |
Knowledge from your people often matters most. The answer to “will this fit my model?” may live in a spec sheet or in the head of your best sales associate. Forward Deployed Engineering is how that knowledge gets captured.
How does the Commerce Context Engine keep data current?
Section titled “How does the Commerce Context Engine keep data current?”The Commerce Context Engine updates different data at different speeds. Shopper behavior on your site arrives continuously through onsite tracking. Your product catalog syncs from your commerce platform on a schedule, every six hours by default. The console shows the sync status and schedule.
The Commerce Context Engine also enriches each product with features extracted by AI from your own product data. Those features help specialist agents answer attribute questions, such as material or fit.
Why does grounded context matter for AI answers?
Section titled “Why does grounded context matter for AI answers?”An AI agent answers well only when it has the right facts. A grounded answer comes from your own data and sources, not from a language model’s general knowledge. The Commerce Context Engine gives every specialist agent the same grounded context, so answers about products, orders and policies come from your systems.
For general questions in your store’s field, such as what a term means, an agent can also draw on the language model’s own knowledge when it is confident. It doesn’t do that for your policies, prices, stock or orders.
Without that context, an agent has two choices. It can guess, or it can say it doesn’t know. The Commerce Context Engine puts the right facts in front of the agent before it answers.
Is the Commerce Context Engine a CDP or a PIM?
Section titled “Is the Commerce Context Engine a CDP or a PIM?”The Commerce Context Engine is neither. A customer data platform unifies customer profiles, usually for marketing. If you already run a CDP, the Commerce Context Engine works alongside it.
A PIM, or product information management system, is where some merchants manage product data before it reaches the storefront. The Commerce Context Engine doesn’t need one. It reads product data from your commerce platform.
What happens to your data?
Section titled “What happens to your data?”Webscale AI does not train language models on your data. To answer a shopper, the AI Shopping Assistant sends the shopper’s message and the relevant store data to the language models it uses. Conversations are kept so your team can review them in the Webscale AI console.
The Commerce Context Engine can also write actions back to your systems where that is set up. For example, Order & Returns Management can start a return in your store when Allow write actions is on.
How do you connect your store?
Section titled “How do you connect your store?”The Commerce Context Engine connects to Shopify, Adobe Commerce and Magento Open Source through built-in connectors. The guides below cover each platform.
- Connect a Shopify store
- Connect a Magento or Adobe Commerce store
- Supported platforms and integrations
Key takeaways
Section titled “Key takeaways”- The Commerce Context Engine is the data layer every Webscale AI product runs on.
- It connects your store, your back office and your people’s expert knowledge into one model-ready context.
- Shopper behavior arrives continuously, while the product catalog syncs on a schedule.
- AI-extracted product features help specialist agents answer attribute questions.
- The Commerce Context Engine is neither a CDP nor a PIM. It needs neither.
- Webscale AI does not train language models on your data. Conversations are kept for your team to review.
Frequently asked questions
- Is the Commerce Context Engine a CDP?
- No. A CDP unifies customer profiles, usually for marketing. The Commerce Context Engine turns your catalog, store data and expert knowledge into context AI agents can answer from. It works alongside a CDP if you already run one.
- Do I need a PIM to use the Commerce Context Engine?
- No. The Commerce Context Engine reads product data from your commerce platform.
- How often does the Commerce Context Engine update?
- Shopper behavior on your site arrives continuously through onsite tracking. Your product catalog syncs on a schedule, every six hours by default. The console shows the sync status and schedule.
- Which commerce platforms does the Commerce Context Engine connect to?
- The Commerce Context Engine connects to Shopify, Adobe Commerce and Magento Open Source through built-in connectors.