Google Merchant Center for Amazon Brands Going DTC: Suspension-Proofing and Feed Optimization

What does Google Merchant Center optimization mean when you come from Amazon?

Google Merchant Center optimization is the work of getting your product feed approved, complete, and structured so Google can sell your catalog across Shopping ads, Performance Max, and free listings. For an Amazon-native brand opening its own store, that splits into three jobs: pass a new-store trust review, translate listing discipline into feed structure, and scale spend as history accumulates.

TL;DR

• Merchant Center is the closest thing DTC has to Amazon's listing infrastructure. At the established brand we managed, 84% of Google spend and 78% of paid revenue ran through the product feed.
• New stores fail the trust review before they fail at selling. Misrepresentation suspensions are about your store's surface, and you can pass deliberately instead of appealing repeatedly.
• Expect a real gap when you launch without brand authority. In a heavily competitive category, our fresh feed clicked through at 0.86% while the established brand's did 1.43%, and month one closed at a 1.46 ROAS. That is the no-authority baseline, not failure.
• Titles are the highest-return field in the feed. Across 57 million impressions, click-through held flat to about 90 characters and fell 11% beyond that.
• Run standard Shopping before Performance Max. Visibility first, automation once the feed has earned it.
• Brand authority is the multiplier. Even the fresh brand's tiny branded demand was its most efficient spend: 14% of budget, 32% of revenue. An established Amazon brand walks in with years of that asset already built.

Your listing skills transfer. The rulebook does not.

Amazon sellers arrive at Google with a real advantage and one dangerous habit. The advantage is feed discipline: you already believe that titles, images, identifiers, and category data decide revenue, which is a lesson most Shopify-native founders learn expensively. The habit is optimizing for an algorithm that rewards accumulation. Amazon lets you stack keywords in backend fields and 200-character titles. Google has no backend field, reads your landing page against your feed, and quietly punishes clutter.

Here is the translation table we walk new clients through.

The mindset shift underneath the table: on Amazon you optimized a listing inside someone else's store. On Google, the feed and the store must agree with each other, because Google checks. That agreement is the entire subject of the next section.

Suspension-proofing: pass the review Amazon never ran

Amazon vetted you once, years ago, and your account history has carried you since. Google re-runs that judgment on your new domain with zero history, and the most common verdict for fresh stores is a suspension for misrepresentation. The name misleads people. It rarely means your products are misrepresented. It means the store, as a whole, has not yet proven it is a real business, and Google's misrepresentation policy casts a deliberately wide net.

The triggers we see most on new Amazon-to-DTC builds:

• Policy pages missing, half-written, or contradicting Merchant Center settings. Returns, shipping, privacy, and terms need to exist, match reality, and match what the feed declares.
• Business identity that does not line up. The store name, the domain, the Merchant Center account name, and the payment descriptor should read as one company.
• No visible way to reach you. A physical address and at least one working contact channel, findable from the footer, not buried.
• Feed data that disagrees with landing pages on price, availability, or shipping cost. Even small drift reads as deception to the review system.
• A large catalog uploaded to a domain that went live last Tuesday. History is a trust input, and you have none.
• Leftover scaffolding: placeholder pages, default theme text, a coming-soon collection that never came.

The order of operations matters more than people expect. Fix the entire list before your first feed submission, because a clean first review is cheap and appeals are not. Re-reviews take days, repeated failed appeals stretch to weeks, and the whole time your launch calendar burns. We treat the product data specification as a pre-flight checklist, not a reference to consult after rejection.

One more Amazon-specific trap: pricing parity. If your DTC price undercuts your live Amazon listing, you invite two problems at once, Amazon suppressing your featured offer and Google flagging mismatches wherever the crawler finds stale prices. Our position on parity pricing is laid out in the Amazon-to-DTC expansion playbook, and it applies unchanged here.

What launching without brand authority actually looks like

We manage Google accounts for two brands in the same product category, at opposite ends of the authority curve, which makes them an honest natural experiment. One is an established brand we ran for two years, past $1.6M in lifetime Google spend, with years of accumulated search demand behind it. The other launched from scratch last October: new store, new feed, a heavily saturated category, and no brand authority anywhere yet. Same team, same playbook, very different starting positions.

Month one closed at a 1.46 blended ROAS. Nobody enjoyed it. Nobody panicked either, because the shape of the next nine months was the plan working:

November and December ride seasonal product-market fit, the post-holiday trough bottoms at 2.05 in February, and spring settles into a stable 2.4 to 2.9 band while monthly spend scales roughly four times. That stabilization under rising spend is the signal that Google has learned the catalog. And it was earned the hard way: every auction in this category runs against incumbents with years of feed history, thousands of ratings, and brand names people already search for. The newcomer pays the same price per click and converts it worse until its own assets accrue.

Two numbers the chart cannot show:

• Shopping click-through ran at 0.86% against the established brand's 1.43%. That gap is not a settings problem. It is what zero ratings, zero history, and unproven price competitiveness look like in a metric.
• Revenue concentrated hard: the top 1% of products produced 63.8% of conversion value, versus 47.3% at the established brand. Young feeds live on their heroes.

Read this chart against an established account's numbers and it looks like failing. It is not. It is what entering a saturated category without brand authority costs, and the honest version of the process deserves spelling out. The first quarter buys data: Google learns the catalog, the store earns its first ratings, the brand starts existing in search at all. The second quarter converts that into a stable band you can scale against. Holding 2.4 to 2.9 while budgets quadruple, in a category where every incumbent has a years-long head start, is a trajectory, not a verdict. The failure mode is quitting at month three, or walking in with efficiency targets borrowed from a brand that had authority you have not built yet. Set targets from your own margin math instead: our break-even ROAS walkthrough exists for exactly that conversation.

The most efficient money in the account was branded search, even for a brand almost nobody was searching for yet: Search took 14% of spend and returned 32% of revenue at a 6.30 ROAS. Now read that number the way an established Amazon brand should. Years of marketplace sales build a pool of people who type your brand into Google every day, and that demand follows you off the platform. The fresh brand's grind above is the baseline without that asset. Arrive with real brand authority and you start several rungs up the ladder: cheap, high-intent branded clicks prove the store and feed the account conversion data while the feed earns its history. Capturing them is the first campaign we turn on, before the feed has even finished review.

Feed optimization, in the order that moves the numbers

Every field in the feed can be improved. They are not equally paid work. This is the order we run, with the data that set it.

Titles first, and unlearn the stuffing. Titles are the single highest-return edit in Merchant Center. Structure beats volume: brand, product type, and the attribute a buyer actually decides on, packed into the first 70 characters because that is roughly what display truncates. Across 57 million Shopping impressions on the established account, click-through held between 1.48% and 1.49% for titles up to 90 characters, then dropped to 1.32% in the 90-to-120 band. An 11% relative penalty for length that Amazon would have rewarded. Front-load, then stop.

Images that survive policy. Your Amazon white-background habit is safe here, and clean product-on-white passes review without drama. What does not: promotional text, watermarks, and badge overlays baked into the main image, which are rejection triggers. Once approved, test lifestyle imagery for Performance Max placements, because feed images travel to surfaces where a sterile white square underperforms.

Identifiers you already own. Amazon forced you to buy legitimate GTINs years ago. Submit them. Correct identifiers connect your products to Google's product graph, which improves matching, comparison placement, and review aggregation. This is a genuine head start over Shopify-native brands that skipped barcodes.

Categorization, both fields. google_product_category places you in Google's taxonomy; product_type is your own free-form tree. Fill both, make product_type granular, and mirror your store's collection logic so the feed and the site tell one story.

Custom labels, because concentration is real. With the top 1% of SKUs producing half to two-thirds of revenue, treating the whole catalog as one bidding blob is a budget leak. We label hero products, seasonal lines, and margin bands, then let campaign structure respect those boundaries. The same audit that sets labels finds the waste tail: on the established account, 204 products had accumulated 30 or more clicks with zero conversion value, a quiet leak of nearly 1% of Shopping spend that pruning ended.

Supplemental feeds for speed. Seasonal title rewrites, promo attributes, label changes: run them through a supplemental feed layered over the Shopify sync instead of editing products one by one. Feed changes should move at campaign speed, not catalog-admin speed.

Ratings, started early because they arrive late. Reviews do not transfer from Amazon, and the click-through gap above is partly that absence made visible. Enroll in Google Customer Reviews at launch and wire your post-purchase flow to feed it. The stars show up months later, which is precisely why the enrollment cannot wait.

Standard Shopping first, Performance Max second

The strong temptation on a new account is to hand everything to Performance Max immediately, because that is what the interface recommends and what most agencies default to. We sequence it deliberately, and the new account currently runs 64.5% of spend through standard Shopping for exactly this reason.

Simona, our co-founder and one of the first certified Google experts in Bulgaria, runs the Google practice and puts it plainly: "Performance Max spends confidently whether or not the feed deserves it. Standard Shopping shows you which queries and which products are earning the budget. A new feed gets that visibility first. Performance Max gets the catalog after we trust it."

The established account shows the destination. By the time we ran it at scale, 71% of spend flowed through Performance Max at a 6.15 ROAS, with standard Shopping holding 13% at 6.89 and brand Search compounding on top at 9.51. Automation is the right endgame. It is a terrible diagnostic tool, which is why it goes second. The fuller comparison of where Google fits beside Meta for a DTC brand lives in our Google Ads versus Meta Ads breakdown.

What the feed pays like once the brand has authority

Two years of the established account, in three numbers: $1.02M of spend, $6.88M of tracked conversion value, a 6.78 blended ROAS. This is the same playbook after a brand has authority: ratings on every hero product, years of feed history, and a name people type into Google unprompted. Set against the fresh brand from the previous section, the two positions look like this:

Underneath those numbers, the structural facts worth copying.

The feed was the business. Performance Max and Shopping together took 84% of spend and produced 78% of paid revenue. Search, mostly brand capture, took 16% of spend at a 9.51 ROAS with a 29% click-through rate. That Search line is what years of brand building pays out: the fresh brand upstream has no version of it yet, and no setting can conjure one. There was no exotic channel mix, just a product feed maintained like infrastructure and a brand name defended like property.

Seasonality was engineered, not endured. In 2024, October through December produced 53% of the year's paid revenue on 41% of its spend, a 6.95 quarterly ROAS against 4.33 across the rest of the year. That did not happen by luck: seasonal hero products entered the feed with corrected titles, labels, and imagery in September, so campaigns scaled into proven data instead of learning during the peak. If your catalog has a season, your feed calendar starts a quarter earlier.

And the long tail earned its keep. Even with heavy concentration at the top, more than half of conversion value came from outside the top 1% of SKUs. Feed hygiene across 21,645 product variants is unglamorous work with no single visible win, which is exactly why most stores skip it and why it compounds for the ones that do not.

Coming from TikTok Shop instead?

Shorter runway, same gauntlet. TikTok-native brands hit the identical new-store trust review, usually with a thinner site and less patience, and every item on the suspension checklist above applies unchanged. The strategic difference is demand shape: TikTok creates spiky, video-driven branded demand instead of Amazon's steady searchable kind. When a video runs, thousands of people search your brand name on Google within hours, and without an approved feed and a brand Search campaign, that surge buys from a reseller or an imitator. The feed work is the same. The urgency is higher. We wrote up the full motion for TikTok Shop brands adding their own store.

The feed is the storefront now

On Amazon, the listing was the business and you treated it that way. Merchant Center deserves the same seriousness, because for a DTC brand on Google, the feed is the storefront: it decides what you appear for, what you pay, and whether the machine trusts you enough to show you at all.

If you are an Amazon brand planning the move, start with our expansion breakdown for Amazon-native brands, then bring us your catalog. We run Google Ads for DTC brands with the feed treated as infrastructure, the suspension checklist cleared before submission, and targets set from your margin math rather than someone else's dashboard. The first review is the cheapest one you will ever pass. Talk to us before you submit it.