Answered plainly, including the parts that are still in progress. If something here matters to you and the answer isn't good enough yet, say so before you buy rather than after.
It doesn't open on its own by default. The shopper opens it — from a launcher you place, or from a trigger you configure (a product page, a category, time on page, exit intent). Auto-open is available and off unless you turn it on. We recommend leaving it off until you've seen how your traffic behaves.
Always. It closes to the launcher and the conversation is preserved, so reopening continues rather than restarts. There is no interstitial the shopper can't get past, and nothing blocks the underlying page.
Inventory and pricing are read from your catalog at the time of the answer, so out-of-stock items aren't recommended as available. Variants are handled where your feed exposes them at variant level. Promotional pricing depends on where the promotion lives — catalog-level price rules are picked up; checkout-time cart promotions are not visible to the agent and it won't claim them.
Where your catalog carries shipping rules, the agent applies them and will ask for a ZIP or region before making a delivery claim. Where it doesn't have that data, it says delivery needs checking rather than assuming.
The agent adds to your existing cart and hands the shopper to your normal checkout. We don't replace, wrap or intercept checkout, and no payment data touches HOLLOU.
This is the part worth reading twice. When retrieval returns nothing relevant, or the validation step can't support a claim against your catalog, the agent says it doesn't know and offers a human. It does not produce a plausible answer to stay useful.
Every one of these is logged as a catalog gap in your console, which over time is one of the more valuable outputs — it's a list of the questions your product data can't currently answer.
A circuit breaker trips and the agent degrades to a clearly worded fallback with a route to support, rather than hanging or erroring at the shopper.
Any website, via one script tag — that's the primary path and it's platform-agnostic. Shopify has a direct catalog sync in production. WooCommerce, Magento and custom stacks are supported through feed or API ingestion, which is production-ready but involves a mapping step during onboarding rather than being one-click.
We'll tell you which category you're in on the first call, and we won't describe a mapping exercise as a one-click install.
Conversations and events are stored on your merchant instance to power your console. Your catalog data is stored to serve retrieval. Personally identifying information in transcripts is masked by default, with reveal-on-demand recorded in an audit log.
No. Your customer data is never used to train external models.
Aggregate counts only — sessions, tokens, cost, health. The platform console holds no message content, and it is architecturally separated from the systems that do. Raw transcripts exist only in your own console, scoped to you.
Data is encrypted in transit and at rest. Our data handling is built to GDPR requirements. Because checkout stays on your platform and no card data ever reaches HOLLOU, we are outside PCI-DSS cardholder-data scope. We are working toward SOC 2 Type II and do not hold a report today — if your procurement process requires a current report, we are not the right fit yet, and we'd rather say so now.
Send us your store URL. We'll index a slice of your catalog and show you the agent answering real questions about your actual products — before you commit to anything.
No catalog credentials needed for the first preview