Availability or price in Merchant Center disagrees with the URL
The auction shows in stock. The page shows sold out, or the opposite. That is eligibility. It is not a collection SEO ticket until the HTML itself is the policy.
Shopping Ads
Shopping Ads to Align Merchant Center Fields with the Product Auction
Shopping ads sell what the feed says: title, GTIN, price, availability, and the landing the auction sends. Merchant Center disapprovals are eligibility. Product groups are how you split that inventory. We work the feed in the auction. We do not run catalog SEO as this H1. We do not treat Performance Max as the whole Shopping job. We do not invent ROAS from a cleaner GTIN.
- Sample Merchant fields
- Split product groups
- Rank disapprovals
- Tie recheck to fetch
Position
The auction sells the feed. The landing has to agree.
A Shopping ad is not a text RSA. It is image, price, merchant name, and the SKU landing. Search intent still has to show on that product page: the shopper who clicked the listing should find the same title, price, and availability. Keyword targeting is one job: the product queries you will buy through this feed, not every synonym merchandising wrote into a collection. When Merchant Center says in stock and the page says otherwise, the auction is selling a lie.
Catalog URL policy
Which products and facets may be documents lives on ecommerce SEO. Feed versus landing mismatches for the auction stay here.
Open ecommerce SEOPerformance Max architecture
Asset groups, signals, brand exclusions, and reporting limits live there. Listing filters can sit in PMax. Merchant eligibility and PLA titles as the ad object stay here.
Open Performance MaxThe PPC family chooser
The hub names whether Shopping is in the program. This page does not reprint that map. Use this URL when the feed is already supposed to be in the auction.
Open the PPC hubA one-time catalog SEO snapshot
An ecommerce SEO audit dates indexation and crawl waste. Merchant disapprovals are this URL. Do not merge the PDFs.
Open ecommerce SEO auditEcommerce SEO owns which product and collection URLs may index. An ecommerce SEO audit is a catalog snapshot. Performance Max can spend on listings with asset groups and weak query reports. The Google Ads account holds conversion actions those Shopping campaigns train on. This URL owns feed fields, product groups, and disapprovals.
We will not treat a theme title tag as a feed title. We will not fold this job into PMax because the UI made Shopping feel old. We will not promise a return from cleaner GTINs. If the catalog HTML is the permit question, open ecommerce SEO. If asset groups are the question, open Performance Max. If conversion actions in Google are fiction, open Google Ads.
When this fits
When the feed and the product page tell two stories
This page fits when Shopping is in the program and the missing object is the feed in the auction. If you needed catalog indexation, PMax asset groups, or Google conversion actions only, start there instead.
GTINs are missing, shared, or belong to another brand
The feed cannot identify the product. Disapprovals pile up. We name GTIN rules as ad fields. We do not invent identifiers to force a listing.
Everything sits in one product group
Margin, stock, and query control need splits. We write groups you can explain. PMax listing filters are not a substitute for this map when you still need Shopping control.
Titles were written for organic snippets and pasted into the feed
Feed titles are the ad object. Search intent on the product page still has to match. Keyword density as an SEO score is not a Shopping grade.
Process
Sample Merchant fields. Split product groups. Rank disapprovals. Tie recheck to fetch.
Four Shopping jobs in this order. We do not start a catalog SEO rewrite in the last step. The point is a feed the auction can sell without lying.
-
Sample Merchant fields
We sample titles, GTINs, price, availability, and landing URLs as Merchant Center sees them. Search intent on the product page is compared to the feed title. Keyword targeting is the product queries you will buy. Semantic keywords belong in the feed title only if they still name the SKU. A spreadsheet export that is a week old is not the live feed.
Outcome A feed sample: mismatches between Merchant Center and the live product URL.
-
Split product groups
We write product groups (or listing splits) that match how you will bid and exclude. Topical relevance of a group to a query cluster is the test. Keyword density inside a feed title is not a metric and not a group score. Catch-all groups that hide waste are named.
Outcome A product-group map you can explain to merchandising.
-
Rank disapprovals
We rank Merchant Center disapprovals and warnings that block or weaken listings. Price, GTIN, and landing mismatches come first. Entity names in product copy belong in schema and AI work. A policy appeal is named only when the feed already tells the truth.
Outcome A ranked disapproval list with an owner for each row.
-
Tie recheck to fetch
We set a recheck after the next feed fetch. Theme and inventory apps can restore a lie. We do not wait for a month of wasted listings. PMax may still spend on the same SKUs. That spend does not close this recheck.
Outcome A recheck date tied to a feed fetch, not to a ranking calendar.
Deliverables
What merchandising and ads can keep after we leave
You leave with a feed sample, a product-group map, a ranked disapproval list, and a recheck tied to a fetch. You do not leave with a catalog SEO retainer, a PMax bible, or a return promise.
Feed sample
Titles, GTINs, price, availability versus live URLs. Dated. Last season export is not this sample.
Product-group map
Splits you will actually use. One catch-all group is named if that is the current state.
Disapproval list
Ranked blockers with an owner. Policy theatre without a true feed is refused.
Recheck rule
When to pull diagnostics after the next fetch. A weekly vanity score is not a recheck.
Handoff lines
When ecommerce SEO, PMax, or Google conversion actions must take the next hour.
Benefits
What a PMax toggle cannot replace
The auction stops selling a stock lie
Feed availability matches the page, or the listing is paused on purpose.
GTINs have a rule
Missing and borrowed identifiers are listed. We do not invent codes.
Groups can be explained
Merchandising sees which SKUs share a bid bucket. Waste stops hiding in All products.
Disapprovals have owners
Diagnostics are not a screenshot in chat. Each row names who fixes the feed field.
Methodology
Who can change a GTIN and a feed title
This works when someone can edit the feed and someone can confirm the live product page. If both are missing, we are reading Merchant Center. We are not holding the auction object.
We start with Merchant Center diagnostics, the live feed, and money SKUs. If any of those is missing, we name it first.
We fetch product HTML. We do not trust the feed UI alone. Does search intent on the product page match the listing? Are the product queries you will buy actually eligible?
You may rename the roles below. You may not leave the feed owner implied. An implied feed is how in stock and sold out coexist.
- Shopping ads lead SEOConsultants.ai writes the feed sample, the group map, the disapproval list, and the recheck rule. We do not become your merchandising team because GTINs were blank.
- Feed owner Can change titles, GTINs, price, and availability. A locked feed with no owner is a finding.
- Whoever owns the product URL Can fix HTML that disagrees with the feed. Catalog indexation still lives on ecommerce SEO.
- Whoever owns Google conversion actions Confirms Shopping campaigns are not training on fiction. That object still lives on Google Ads.
- Leadership Accepts that cleaner GTINs are not a return promise. An unread disapproval list is not a failure of this page. A catalog SEO reprint would be.
First working session
Start with one SKU the feed and the page disagree on
The first hour is one money product: feed row versus live URL versus Merchant diagnosis. It is not a tour of PMax asset groups.
We pick a SKU that spends. We open the feed row, the product page, and the diagnostic. If availability or price disagrees, that is the job. If the real question is faceted URLs, that is ecommerce SEO. If the real question is asset groups, that is Performance Max. If the real question is Google conversion actions, that is Google Ads.
Then we list product groups and a sample of disapprovals. The catch-all group and the GTIN gaps are usually the job. A Merchant Center dashboard that looks green while the page says sold out is not evidence.
We leave with a first mismatch list or a stop. A list names the field, the owner, and the recheck after the next fetch. A stop names what is missing: feed access, a landing that is actually a collection SEO problem, or a job that is actually PMax. Both are outcomes. A catalog essay is not.
AI search
A GTIN is not an AI Overview
Honest product HTML and matching feed facts can help later extraction. This page cannot sell AI Overviews, a chat mention, or LLM visibility as a score.
A cleaner Shopping feed will not land products in Google AI Overviews (AIO). If feed and page disagree, generative search has two prices to lift. That is ours to tidy as eligibility. If the product page already tells the truth and models still invent a GTIN or a stock state, that is an AI search audit. A program across search UIs is AI SEO.
The product URL has to print the facts
AI search needs a page it can fetch. A feed-only title that never appears on the page does not help. We do not sell Overview presence.
SKU names should match the thing you sell
We check product names in HTML and feed. That is entity optimization as a Shopping input. It is not a Knowledge Graph project from this page.
Measuring model answers is another URL
LLM visibility needs a sentence a model can quote, then measurement on LLM SEO. A feed sample can show a price lie. It does not sample answers.
Schema
Product markup must match the feed and the page.
Product JSON-LD helps a machine read a SKU the page already states. This work can flag a mismatch with Merchant Center. It is not a schema campaign as the whole job.
Price and availability in markup must match the feed
If schema, feed, and HTML disagree, extractors guess. We name the mismatch. We do not add Offer stuffing to chase listings.
No fake FAQ on a product URL to host more queries
Keyword targeting is the product queries you will buy. FAQ blocks that exist only to host terms are refused.
Extra types do not buy a citation
Markup will not force generative search to cite the SKU. Eligibility is not a purchased mention.
Who we work with
The person who can edit the feed and the product URL
Someone has to fix a GTIN. We work with that person. We do not replace them with a PMax toggle.
Feed owner
Ships titles, GTINs, price, availability. Rechecks after the next fetch.
Product URL owner
Keeps HTML honest. Holds catalog SEO on its own page when the permit is indexation.
Whoever already runs PMax or Google Ads
Can keep going once listing filters and conversion actions are named. If they try to run this as a catalog SEO retainer, two jobs collide.
Questions
Frequently asked questions
How is this different from ecommerce SEO?
Ecommerce SEO owns which catalog URLs may be documents: facets, product indexation, collection rules. This page owns the feed in the auction: titles, GTINs, disapprovals, product groups. A Merchant Center in-stock flag that disagrees with the product page is this job. A faceted URL policy is that job.
Is Performance Max the whole Shopping job?
No. Listing-group filters can live in PMax. Merchant eligibility, GTIN and title as the ad object, and product groups for query-level Shopping control stay here. PMax asset groups and reporting limits live on Performance Max.
Are titles and GTINs on-page SEO?
No. Here they are feed fields that become the Shopping ad. Organic title tags stay on ecommerce SEO or on-page SEO. Keyword targeting is the product queries you will buy through the feed, not a title-tag OS.
Do you set a return target as the Shopping product?
No. Margin can exist as a feed label. We will not print a return promise or invent ROAS from a cleaner GTIN. Disapprovals and mismatches are the object.
Will a cleaner Shopping feed get us into AI Overviews?
No. Honest product HTML and matching feed facts can help later extraction. This work does not buy a place in Google AI Overviews (AIO), a generative-search mention, or an LLM visibility score. If models invent a price or a GTIN you do not sell, that measurement sits on an AI search audit and on LLM SEO.
What if Merchant Center looks green and the page says sold out?
That is this job. The auction is selling a lie. We sample the live feed against the product URL. We do not treat a green diagnostic as evidence until the page agrees.
What do you need to open Merchant Center?
The Merchant Center ID, a sample of disapprovals, and one product URL that disagrees with the feed. A homepage walkthrough is not a feed brief.
The feed in the auction
Does Merchant Center still say in stock while the product page says sold out?
Share the Merchant Center ID, a sample of disapprovals, and the product URL that disagrees with the feed. We will say if Shopping eligibility fits, or whether ecommerce SEO, Performance Max, or the Google Ads account should go first.
- Feed fields matched to the landing
- Disapprovals ranked with owners
- Product groups you can explain
- No invented ROAS from GTINs
Merchant fields, groups, disapprovals, fetch recheck. Not catalog SEO.