• Shopware Gold Partner & Shopify Partner

Chapter 8

Trust signals that influence AI answers

Table of contents

Chapter 5 asked: can a machine understand your product data at all? This chapter asks the follow-up question: does it have enough information to recommend your store with confidence?

This marks the transition from dimension one to dimensions two and three of the AI readiness model: trust and conversion. A store can be flawlessly set up on dimension one and still fail here.

For a machine, a well-told brand story alone is not enough. In concrete purchase decisions, what counts above all is information that can be captured unambiguously: delivery time, shipping costs, return period, reviews and availability. Chapter 4 covered the external side of this question – what others say about you. This chapter is about the internal side: which trust data your own store provides in machine-readable form.

Which two fields offer the best ratio of effort to impact?

Chapter 2 mentioned that Shopware has included shipping information in the product schema since version 6.7.9.0. Return policies are a different matter – and this is exactly where the biggest open gap lies in many stores.

  • shippingDetails: shipping costs, destination region and ideally delivery time.

  • hasMerchantReturnPolicy: return category, period, costs and explicitly the country the policy applies to.

Both have evolved from "recommended" to effectively expected. Google still lists them as recommended in its documentation on merchant listings and points out that shipping and return information can improve how products are displayed in results. Important for the effort: the information does not have to be on every single product. Google also accepts return and shipping policies once, centrally, at organization level – for stores with uniform conditions, that is the simplest solution.

The return information belongs in the schema even if you do not accept returns at all. There is a separate category for that. "No returns" is information. No information is a gap. Google requires a clearly communicated return policy on the website anyway; missing information can lead to rejection in Merchant Center.

Also important: the country the return policy applies to is stated explicitly (applicableCountry). For a store that ships to Germany, Austria and Switzerland, this is not a detail but the difference between usable and unclear information. If a return goes to a different address – for example a returns warehouse abroad – that country belongs in it too (returnPolicyCountry).

Which detour via Merchant Center is often overlooked?

If you run a well-maintained Merchant Center account, you can store shipping and return conditions centrally there – and now also directly in Search Console. This makes particular sense for information that changes more often. A fixed order of precedence applies: if the conditions are in several places, Google uses the settings from Merchant Center or Search Console rather than the markup on the page. Consistency is therefore decisive: Google requires the information in Merchant Center to match the information in the store. Diverging shipping costs can lead to warnings or even rejection.

What counts technically for reviews?

Chapter 4 treated reviews as external authority. On your own site, there is an additional requirement that many stores fail: reviews in the schema must also be visible on the page. Google explicitly requires structured product data to match the information a user sees on the landing page. Schema and visible content must not contradict each other. This is the same consistency principle as in Chapter 5 – here with immediate consequences.

What do actual delivery times have to do with your delivery promise?

A point that is rarely discussed: Google can now calculate actual delivery times itself. Merchants who share their tracking data with Merchant Center enable Google to determine the real delivery speed per region. If it is faster than the stated delivery time, Google can display the shorter time. Stated and actual delivery times thus become comparable.

The practical consequence is clear: delivery times become a verifiable value instead of a claim. A realistic delivery date that can be kept is therefore more valuable in the long run than an optimistic one that is broken more often – and if you deliver reliably fast, you can now make this advantage visible.

Our assessment

Most stores do not have a knowledge problem here, but an implementation problem. Return policies and delivery times are almost always somewhere on the site – in body text, in the terms and conditions, in the footer. For a person, that is enough. For a machine, the same information is much more reliably usable if it is also in the structured fields intended for it.

The effort to change this is manageable: two schema blocks or correctly maintained Merchant Center settings. Measured against the data work that Chapter 5 calls for, this is the cheapest improvement in the entire guide.

What you can check in ten minutes today

  1. Open a product page and check with a schema testing tool whether shipping and return information is actually present as structured fields – not just as text somewhere on the page.

  2. Check whether the marked-up reviews are also visible on the page.

  3. Compare the delivery date shown at checkout with your actual average delivery time over the last few weeks.

Frequently asked questions

Yes. There is a separate category for this case in hasMerchantReturnPolicy. "No returns" is information – missing information is a gap and can lead to rejection in Merchant Center.

Shipping information yes, since version 6.7.9.0 with JSON-LD enabled. You usually have to add return policies yourself – either in the schema or centrally in Google Merchant Center.