264: What Amazon Strategies Are Working Today With Brad Moss

Podcast: Download (Duration: 45:57 — 52.9MB)

Amazon organic rank is driven by sales velocity on a keyword, so the most reliable way to rank is to run sponsored product ads against your target keywords and track ranking movement rather than return on ad spend. Brad Moss, the former head of Seller Central at Amazon, calls this the most authentic way to grow organic position because you are using Amazon’s own systems to do it.

Brad now runs productlabs.net, where his team takes over the entire Amazon channel for seven and eight figure sellers. He has spoken at my conference, the Sellers Summit, three years running, and he is my go-to person on how Amazon actually works from the inside.

Below is his keyword volume math, the four inputs that actually control conversion rate, the metric he uses instead of ACoS, and what happened when Amazon cut off 10,000 vendors without warning.

Key takeaways

• Amazon SEO is sales-driven: the top result for a keyword is generally the product with the most sales on that keyword right now.
• The top three listings on a keyword capture 40% to 60% of that keyword’s traffic combined.
• Amazon conversion rates average around 10%, versus roughly 3% for general ecommerce.
• Reviews are one of four conversion inputs, alongside price, images, and listing text. Brad took a product to a $1 million run rate with zero reviews.
• Track total ad spend divided by total sales, not ACoS, because good ad spend also drives organic sales that ACoS never credits.
• Amazon stopped fulfilling purchase orders for over 10,000 vendors with no warning, which is the case for building on Seller Central instead of Vendor Central.
• Enhanced Brand Content produced minimal lift in Brad’s client data, largely because over 60% of shoppers browse on mobile where it gets squashed.
• There are over 300,000 manufacturers in the United States, which is an underused supply chain option against China-based sellers.

Why Amazon cut off 10,000 vendors without warning

Amazon stopped fulfilling purchase orders for over 10,000 vendors with no advance notice, and affected suppliers received either a vague system-error message or a flat “we are not reordering from you.” Brad notes that Amazon does this roughly every year around late February or early March, and the prior year’s version was shutting down Vendor Express.

Amazon is acting as a retailer here, so it has every right to stop ordering. That does not help a business whose entire revenue depended on those purchase orders.

The recourse is shifting to third-party selling on Seller Central. Once Amazon sells through the last of its inventory, a vendor with a Seller Central account can take control of those ASINs and listings.

Some of Brad’s clients buy their remaining inventory back from Amazon to accelerate that transition. You cannot change anything about a listing while Amazon is still selling your product through Vendor Central.

Vendor Central vs Seller Central: which model gives you control

Vendor Central is autopilot and Seller Central is hands on the controls. Brad’s analogy is that Vendor Central sets the plane on autopilot while Seller Central puts you in the cockpit with every knob available.

The vendor model serves very large companies well and serves mid-sized and small companies poorly. Brad has friends at Amazon managing 500,000 SKUs or more, so no individual seller’s few thousand SKUs get meaningful attention.

That means predictive reordering systems run your business. If you know spring is coming and your sales will jump 30%, you have no way to make Amazon order 30% more.

For a seller running Prime, the customer sees no difference between first-party and third-party. Brad’s own preference is third-party for the control, and my recommendation for this audience is the same.

How Amazon organic ranking actually works

Amazon’s ranking algorithm is fundamentally about sales on a keyword, which makes it completely different from Google’s. Whoever is number one for “mechanical pencil” is generally whoever has the most current sales on that keyword.

That creates a direct loop between advertising and organic position. Sales driven by your sponsored product ads on a keyword push you up the organic ranking for that same keyword.

Brad’s recommendation is to build a list of your 50 most important keywords, put ad budget behind them, and track ranking movement as the primary success metric. Return on that specific ad matters less than the organic position you are buying.

He is explicit that this beats the various ranking hacks circulating in the seller community. Using Amazon’s own advertising systems is the approach Amazon actually rewards.

Amazon does not release data on which keywords convert organically. Brad’s team built internal algorithms to back into it, and he describes the math as genuinely complicated.

The Amazon keyword math: estimating sales before you launch

Brad’s team estimates unit sales from search volume, position, and conversion rate. The worked example runs like this.

A real product has 25 to 50 keywords worth running this on, so you stack the estimates. The 10% conversion figure is an Amazon-wide average and varies by category.

For comparison, general ecommerce conversion sits closer to 3%. Amazon converts roughly three times better because the shopper arrives already intending to buy.

This math is also how Brad sets expectations. Sellers regularly arrive wanting to do a million dollars on a product where the search volume supports $500,000, and the research says so before any inventory gets ordered.

How Brad scores whether a category is too competitive

Brad’s team builds an internal competitiveness score from the top sellers in a category, their estimated keyword search volume, their review counts, and how well developed their listings are. Weak listings held by top sellers are an opportunity signal.

The market has shifted from land grab to real estate development. Putting a product up and letting it sell stopped working, and the sophisticated sellers who study their analytics and their competition are the ones surviving.

Knowing your bottom line rather than just your top line is part of that sophistication. So is knowing the ceiling on each product before you commit.

The four inputs that control your Amazon conversion rate

Brad treats conversion rate as the product of four inputs: reviews, price, images, and listing text. Reviews are one of four, not the whole game.

His proof is a client that launched a brand new product into an old category dominated by established brands and reached a $1 million run rate with zero reviews on the product. The incumbents had two or three photos and thin descriptions.

Brad’s team came in with six to nine images and a description written to the actual consumer need. The product was also American made in a category where buyers cared, which competitors sourcing overseas could not match.

Because those inputs were strong, the listing needed very little marketing spend to take off. Brad raises this specifically to argue against the idea that there is a silver bullet.

Do Amazon backend keyword fields matter for ranking?

Backend fields like intended use and subject matter have very little effect on keyword rankings. They feed Amazon’s internal category and use-case mapping rather than search.

The vast majority of shoppers search by keyword, so your core keyword fields carry the weight. Brad frames the rest of the backend fields as the final 2% to 5% of listing completeness, which can take as long as everything else combined.

He does not dismiss them entirely. Occasionally a seller fills in a field and sees growth, and across his client base the core keywords are what consistently matter.

Why ACoS is the wrong metric for your Amazon business

Use total ad spend divided by total sales instead of ACoS when evaluating your business. ACoS only counts sales attributed directly to an ad, which ignores the organic sales those ads generated by pushing your keyword ranking up.

A 50% ACoS looks alarming and can be perfectly healthy if the spend is moving you up the organic rankings on the right keywords. The organic sales that follow never show up in the ACoS calculation.

Brad’s simple version: if your total marketing budget was $30,000 and total sales were $200,000, your ad spend over sales is 15%. That is the number that comes out of your margin.

ACoS still has a job. It is the right metric for comparing one campaign against another, and the wrong metric for judging a product or a business.

Counting giveaways as marketing spend

Brad also folds product giveaways into the marketing budget. Giving away $10,000 of product at a 10% margin means spending $9,000 on advertising.

The right question is whether that $9,000 buys more ranking as a giveaway or as sponsored product ads on the same keyword. Treating it as free promotion hides the real cost.

The complication is that not all marketing feeds Amazon rankings equally. Amazon ads bid directly on keywords and those sales move your rank, while Facebook spend may not attribute cleanly at all.

Brad’s practical answer is to look at it three ways: campaign by campaign with ACoS, product by product with ad spend over sales, and business-wide with everything included.

External traffic strategies for Amazon sellers

Brad’s most effective external strategy is building Facebook lookalike audiences from Amazon shipping report data. You download your shipping reports, extract who you shipped to, and upload that to Facebook to find similar people.

The efficiency gain was dramatic in his testing. Click costs dropped from around $1.20 to 40 cents, and on some campaigns to 20 cents.

You are not targeting your actual Amazon customers with ads, so this stays clear of the terms of service problem. Facebook’s own lookalike system does the work from a seed list.

Which customers to seed a lookalike audience with

Seed your lookalike from repeat purchasers rather than everyone who has ever ordered. A customer who bought twice is a much stronger signal of who your real buyer is.

Brad’s team also layers in geographic analysis to see which regions over-index for a brand. Two brands at the same $20 price point behaved completely differently in Florida, with jewelry over-indexing and athletic socks under-indexing.

That data has uses beyond ad targeting. An aquatics brand discovered it was selling heavily in the Great Lakes region and weakly on the coasts, went after a regional license aimed at lake customers, and doubled down on that market.

Geographic data is also useful in conversations with retail buyers. It tells you where your brand already has demonstrated pull.

How to compete with Chinese sellers on Amazon

Brad’s answer is to compete on customer knowledge and branding rather than on price, since price is only one of four conversion inputs. If you live in your customer’s country and speak the language, you can market and brand to them in ways an overseas competitor cannot.

His example is a brand selling the identical product Amazon itself sells, at $16 against Amazon’s $8, and outselling Amazon. Better branding, better reviews, and a better detail page carried a 2x price premium.

That brand is Amazon-only and it wins on listing quality plus unusually aggressive customer service. When a buyer had a problem, they responded the same day, sent free product, and got on the phone.

Brad’s framing is that customer service may or may not be your differentiator depending on your category. If everyone in your category already does an A+ job, find a different edge.

Domestic manufacturing as a competitive advantage

There are over 300,000 manufacturers in the United States, which most sellers never investigate because Alibaba makes overseas sourcing so easy. Domestic sourcing is a real supply chain option worth evaluating.

It also feeds directly into the conversion inputs. In the zero-review case study, “made in America” was a genuine differentiator in a category where buyers cared about it.

The underlying question Brad wants sellers to answer is how their business differentiates from everyone else’s. Launching a me-too version of a trending product is moving inventory, not building a brand.

Does Enhanced Brand Content increase Amazon conversions?

Enhanced Brand Content produced minimal lift across Brad’s client base, with no large changes in conversion rate. His read is that it adds a little brand confidence and not much more.

The reason is mobile. Over 60% of shoppers view products on mobile devices, where Enhanced Brand Content gets compressed into one image and a small snippet.

What actually does the work on mobile is the hero image and the first two or three photos. Most sales still complete on desktop after a mobile browse, so the mobile experience is where the decision starts.

Amazon’s private label brands and how sellers respond

Brad is openly frustrated with Amazon promoting its own private label brands through placements no third-party seller can access. Search for a name brand and Amazon may serve a full-width interstitial pushing its own version at half the price.

He has multiple examples of Amazon using capabilities only Amazon has to squeeze out other sellers. He is candid that he does not know what to do about it.

His counterweight is that a large part of Amazon genuinely wants sellers to succeed, and features do eventually reach third-party sellers. The pattern is lag: an ad placement Amazon used a year ago opens up to sellers later.

Newer seller tools do provide real value. The advertising portal now reports how many buyers are new to your brand, and placement bidding lets you bid extra for top-of-search, product pages, or the rest of search results.

Brad’s advice on placement bidding is to compare ACoS across the three placements and balance them. Pushing everything to first page placement is short-sighted.

You can find Brad at productlabs.net, where his team runs full Amazon channel operations with long-term strategic plans for seven and eight figure sellers.

Frequently asked questions

How does Amazon decide which product ranks first?

Amazon’s organic ranking is primarily driven by sales on a given keyword, so the top result is generally the product currently selling the most for that search term. This makes it fundamentally different from Google’s algorithm.

What is a good conversion rate on Amazon?

Around 10% is the general benchmark across Amazon categories, according to Brad Moss. General ecommerce sites convert closer to 3%.

How much traffic do the top three Amazon listings get?

The top three listings on a keyword capture 40% to 60% of that keyword’s total traffic combined. A product sitting at position three can expect roughly 10% to 20% of the keyword’s search volume.

Do you need reviews to sell on Amazon?

No. Reviews are one of four conversion inputs alongside price, images, and listing text, and Brad Moss took a client product to a $1 million run rate with zero reviews by dramatically outclassing incumbents on images and description quality.

Should you use ACoS or total ad spend over sales?

Use ACoS to compare campaigns against each other, and use total ad spend divided by total sales to evaluate a product or the business. ACoS misses the organic sales your advertising generates by improving keyword rank.

Do Amazon backend keyword fields affect rankings?

Very little. Fields like intended use and subject matter feed Amazon’s internal category mapping, while the core keyword fields are what drive search rankings.

Does Enhanced Brand Content improve Amazon conversions?

Brad Moss has seen minimal lift from Enhanced Brand Content across his clients. Over 60% of shoppers browse on mobile, where the content is compressed to a single image and a short snippet.

Is Seller Central better than Vendor Central?

For most small and mid-sized businesses, yes. Vendor Central runs on Amazon’s automated reordering with no seller control, while Seller Central gives you direct control over pricing, inventory, listings, and advertising.