Conflicting return terms corrected before a successful Merchant Center review

A later re-audit recorded that the material contradictions had been addressed in the reviewed scope. Its recommendation changed from not requesting review to being ready to request review, subject to confirming the evidence behind active promotional claims.

Google subsequently confirmed that the Misrepresentation review was complete and the issue no longer appeared in the same Merchant Center account. The audit and the review-completion email match by account ID; the outcome is supported by Google's notice, not inferred from our readiness recommendation.

The starting point

The initial review examined the public website. Business identity and public company information were consistent in that scope, and a representative product could be added to the cart and taken through to checkout.

Those checks did not resolve the more specific questions about the customer's return entitlement, the policy information exposed in structured data and the terms behind the product's coverage claims.

Merchant Center settings and product-feed evidence were not supplied for that initial audit. The account side was therefore unverified at that stage, rather than assumed to match the website.

Three findings that needed attention

1. The headline return promise was broader than the written terms

The sampled product page promised a full refund and free returns or collection. The written refund policy limited free collection and change-of-mind returns to eligible orders in specified regions.

The audit record also included the merchant's intended service area, which was broader than the regions named in that policy, with an exclusion. That created two questions: what return offer was actually available, and where was it being communicated inaccurately?

The recommendation was to make the real geography, eligibility and charges clear and consistent across the product promise, refund policy and related customer-facing wording.

2. Structured return information disagreed with the visible policy

The site's machine-readable return-policy information pointed to a page that returned a 404. It also included a return-fee amount that was not clearly reconciled with the visible terms. The representative product's structured data contained a brand-spelling inconsistency.

This was more than a missing policy page: a policy existed, but different versions of its location and terms were exposed to readers and systems.

The audit recommended correcting the policy destination and checking the fee and identity fields against the real offer. Adding a visible page would not, by itself, correct those other sources.

3. Warranty and insurance wording lacked accessible full terms

The product page presented a five-year coverage claim using premium-insurance wording. Other policy text referred to a warranty policy, but complete linked terms were not visible in the reviewed navigation or checkout links.

The initial audit treated this as a potential transparency risk. It did not establish that a valid warranty did not exist; it recorded that the full coverage terms could not be found in the reviewed locations.

The required next step was to make the applicable coverage, exclusions and claim process accessible, then ensure the headline description accurately reflected those terms.

What still needed confirmation

The follow-up found no material unresolved identity, policy, product-data, cart or checkout contradiction within the refreshed checks. It nevertheless recorded two remaining points.

First, the store used reference prices, countdowns and urgency statements. A live visit could show those claims, but could not establish the underlying price history, stock evidence or campaign schedule. The merchant needed to confirm that supporting evidence and remove or revise unsupported claims before requesting review.

Second, one terms statement described the warranty as applying to eligible products, while the dedicated policy and sampled product pages described a broader scope. No conflicting exclusion was identified, but aligning that wording was recommended as a small clarification.

The record also did not independently test fulfilment, customer service, actual return handling or warranty performance. Checkout testing stopped before customer details, final address-specific delivery calculation and payment.

Why this matters for an agency

The useful output was a set of specific comparisons: a customer promise against a policy, a visible policy against structured data, and a website offer against account evidence.

That gives an implementation team something to investigate and correct. It also gives the next reviewer a defined question to check, rather than an open-ended instruction to make the store look more trustworthy.

A diagnostic report should distinguish demonstrated inconsistencies, possible risks and missing evidence. After changes, those same findings need a fresh check; the original report alone cannot show that the corrections are live.

This historical case included direct account inspection in its later stage. Semstar's current white-label agency diagnostic uses evidence supplied by the agency instead. Your agency retains the client relationship, implements changes and submits review requests; missing evidence limits what the diagnostic can verify.

If your review has already failed, use this guide to organise the next investigation, including the notice, evidence gaps and verification of changes.

Discuss a similar case

Managing a retailer's suspended account? Discuss a white-label diagnostic for your agency. Scope and pricing are agreed after a personal conversation.

Working on your own store? Request a suspension assessment.

See the redacted audit example to review the report structure. Access is free after entering your email; the sample covers broader recovery work than the agency diagnostic.