Frequently asked questions
What the product does, what it does not do, and how it does it.
What exactly does AgentReadyGo test?
We send real autonomous AI agents, driving an actual browser, to run shopping missions on your store: find a product, understand its price and availability, select a variant, add it to the cart, determine shipping costs, and reach the checkout funnel.
The score does not come from a technical checklist but from what those journeys actually achieved.
Do you place real orders?
Never. Agents always stop before payment, and that limit is enforced in code, not just in the model's instructions: navigation to payment-processor domains is blocked at the browser level, and any click on a button like "Pay", "Place your order" or "Complete purchase" is refused.
No card data, no password, and no personal data is ever entered.
Do you create accounts on my site?
No. Account creation is blocked. If your journey requires an account before showing a price or a shipping cost, that is precisely what we document as a major friction for AI shoppers.
How much load does this put on my server?
One audit is a few dozen page views — the equivalent of one attentive human visitor. Ad and tracking requests are blocked, and our agents identify themselves with a dedicated user-agent containing "AgentReadyGo".
What happens if my site blocks bots?
The report says so explicitly and no score is assigned. It names the address we requested and the response we received — an HTTP 403, for instance — and stops there. We do not grade a store we never got inside.
That is not the same as an audit that ran. When our agents do get in but cannot verify an essential shopping step — adding an item to the basket, for instance — the audit is carried out, every dimension we could measure is published, and only the single overall figure is withheld. The report says which step could not be verified.
What it does not do is tell you why. A refusal on our request does not establish what caused it, and it does not establish that other visitors or shopping agents would meet the same restriction. If the audit could not be carried out, email us and we refund you.
How do you know the agent really reached checkout?
We do not take its word for it. Every stage the agent declares is cross-checked against what the browser actually recorded. Loading an address is not evidence of what happened there: reaching /cart does not establish that an item was added — what does is the basket itself saying so, a quantity that went up, or a confirmation the page displayed. Without one of those, the report says the step could not be verified instead of announcing a success.
The same principle applies to the problems we find. One is marked "confirmed" only for what the observation actually shows — the page it was seen on, and no wider. Whatever a finding claims beyond that is published separately, as an observation to confirm, and counts for nothing in the score.
Do I need to install a plugin or grant access?
No. The product works only on your public storefront: URL → payment → audit → report. No integration, no admin access, no access to internal data, and no account to create.
How is the score calculated?
Eight dimensions are measured. Each aggregates the result of the stages the agents actually cleared, minus a penalty proportional to the severity of the frictions found. Checkout and cart weigh the most.
Technical checks (robots.txt, sitemap, structured data) count for a minority share, and only on the Discoverability and Product Understanding dimensions. No score is produced when the audit could not be carried out — and none is published either when the audit ran but an essential shopping step could not be verified, because a single figure would then summarise a purchase nobody watched.
What does an audit cost, and is there a subscription?
$9 for one store, one audit and one report. One-time payment: no subscription is started and there is nothing to cancel.
The report stays available permanently through the secret link sent to you after payment.
How is this different from SEO or GEO?
SEO is about being found by a human who searches. GEO is about being recommended by an AI. AgentReadyGo measures the next step: whether the agent actually succeeds in getting to the purchase.
Which platforms are supported?
All of them, as long as the store is publicly reachable: Shopify, WooCommerce, BigCommerce, Magento / Adobe Commerce, Salesforce Commerce Cloud, Squarespace, Wix, PrestaShop, Shopware, and proprietary platforms.
How long does an audit take?
Expect five to fifteen minutes for the three journeys. You can close the page: the report stays at the same address, and you get the link by email.
Can I have a site tested that I do not own?
The audit is meant for the site owner or their agency. We only access public pages, at a volume comparable to one attentive visitor, and our agents identify themselves clearly in their user-agent.