How to Rank a New Product on Amazon When You’re Starting on Page 10

You did everything the courses told you to. The listing is live, the photos are good, inventory checked in at the fulfillment center. Then you type your main keyword into Amazon and start scrolling. Page three. Page six. Somewhere around page ten, there it is, your product, sitting next to a phone case that has nothing to do with your category.

If you’re trying to work out how to rank a new product on Amazon from that position, this guide is for you. We’ll explain why new listings start so far back, what Amazon has actually said about it (more than most sellers realize), and a practical order of operations for climbing. We’ll also be clear about what we don’t know, because Amazon doesn’t publish its ranking formula, and anyone who claims to have the full recipe is guessing.

Why your new product starts on page 10

Here’s the uncomfortable truth: a brand-new ASIN (Amazon Standard Identification Number, the product’s catalog ID) is missing the data Amazon’s search relies on most.

This isn’t just seller-forum theory. Amazon’s own researchers have written about it. A 2020 paper from Amazon scientists, presented at The Web Conference, opens with a blunt line: new products in e-commerce platforms suffer from cold start. A newer Amazon paper, published on arXiv in February 2026, explains why. Ranking models lean heavily on behavioral features such as click-through rate, add-to-carts, purchases and ratings. A new product has none of that history, which, in the authors’ words, leads to suboptimal ranking decisions.

So page ten isn’t a judgment on your product. It’s the absence of evidence.

What Amazon does about cold start

The same research also shows Amazon actively tries to give new products a fair look. The 2020 paper describes predicting “priors” (educated starting estimates) for new products’ behavioral features, and reports more new-product impressions in an A/B test across 140 million queries. A 2022 Amazon paper tested a related approach on 50 million queries and reported a 13.53% lift in new-product impressions and an 11.14% lift in new-product purchases. The 2026 paper describes borrowing signals from similar “substitute” products and says that model has been in production since 2025.

Two honest caveats. Research papers describe experiments, not necessarily the exact system running today (the 2026 paper is the exception, since it says it’s live). And none of them gives a fixed timeline. That’s why we treat the popular idea of a guaranteed launch boost with caution, which we unpack in the Amazon “honeymoon period”: what’s real.

The practical takeaway is simple. Amazon gives new products some exposure based on relevance and similar products, then lets real shopper behavior decide where they settle. Your job is to make that early exposure convert.

How to rank a new product on Amazon: what actually moves it

Amazon has never published a list of ranking factors, and it has never officially named an “A10” algorithm, whatever you’ve read. What sellers and tool makers broadly agree on, and what Amazon’s published research supports, comes down to two buckets:

• Relevance: does your listing clearly match what the shopper searched? This is driven by your title, bullets, description, backend search terms and attributes.
• Performance: when shown for that search, do shoppers click, add to cart and buy? Do they stay happy afterwards (reviews, returns)?

Price, availability and fulfillment speed sit underneath both. A product that’s out of stock or priced well above its neighbors struggles to convert, so it struggles to rank.

Notice what’s missing from that list: tricks. Every durable ranking we’ve seen across 6,000+ launches came from a product that was relevant for a keyword and then converted on it. The clever stuff tends to wear off.

Step 1: Fix relevance before you spend a dollar

Ads and deals amplify whatever your listing already is. If it’s unclear, you’re paying to amplify confusion.

Start with the basics:

• Title: lead with what the product is, in the words shoppers use. Brand, product type, key attribute, size or count. Readable beats stuffed.
• Bullets: answer the questions a shopper would ask in the first ten seconds. What does it do, who’s it for, what makes it different, what’s in the box.
• Backend search terms: cover synonyms, alternate spellings and related terms that don’t fit naturally in visible copy. Don’t repeat what’s already in the title, and don’t add competitor brand names.
• Attributes and category: fill in every relevant attribute field and check the product sits in the right browse node. Wrong category placement is a quiet killer.
• Images and A+ content: these don’t add keywords, but they drive conversion, which drives everything else.

A quick aside on AI shopping. On May 13, 2026, Amazon renamed its Rufus assistant to Alexa for Shopping, according to Amazon’s own announcement page. Amazon told investors in April 2026 that Rufus monthly active users were up 115% year over year. We don’t know exactly how the assistant chooses which products to mention, and Amazon hasn’t said. But a listing that plainly answers real shopper questions in its bullets, images and A+ seems like a reasonable bet for both classic search and conversational search.

Then check you’re actually indexed. Search your main keyword together with your ASIN or brand name, or use your rank-tracking tool. If you don’t appear at all, ads won’t save you. That’s an indexing problem, and it needs fixing first.

Step 2: Pick keywords you can actually win

This is where most launches go wrong. Sellers aim at the biggest, broadest keyword in the category on day one and spend weeks burning budget against listings with thousands of reviews.

A better approach is to build a keyword ladder:

1. Long-tail, high-intent terms (lower volume, very specific, close to your exact product). You can realistically convert here early.
2. Mid-volume terms that share words with the long tail.
3. The head term you ultimately want.

Why does this work? Because sales on a specific term seem to tell Amazon you’re relevant for the related, broader terms too. It also protects your conversion rate. A shopper who searched “bamboo cutting board with juice groove” is far more likely to buy your bamboo board with a juice groove than someone who typed “cutting board”.

Picture a supplement brand launching a magnesium glycinate capsule. Going straight after “magnesium” puts it up against huge, established listings. Starting with “magnesium glycinate 400mg capsules” and similar specific terms gives it a chance to convert, earn reviews and build a sales history it can carry up the ladder.

If you’re Brand Registered, the Search Query Performance report in Brand Analytics shows impressions, clicks, cart adds and purchases for search terms tied to your brand, with your share of each. It’s the clearest view you’ll get of where you’re winning and where you’re just being seen.

Step 3: Get enough reviews to stop losing the click

Do you need reviews to rank? Not strictly. But zero reviews next to competitors with hundreds hurts click-through and conversion, which hurts rank.

The safe, Amazon-approved routes for a new product are limited, and that’s on purpose:

• Amazon Vine. At the time of writing, Amazon’s US Vine page lists enrollment fees per parent product of $0 for up to 2 units, $75 for 3–10 units and $200 for 11–30 units. Products need fewer than 30 reviews, must be fulfilled by Amazon and the brand must be in Brand Registry. Amazon charges 30 days after enrollment, and only once the first review is published. Vine reviewers give honest opinions, so a weak product gets weak reviews.
• The Request a Review button (or an approved tool that automates it) for every eligible order.
• Good packaging and a product that matches the listing, which matters more than any insert.

What you can’t do: offer anything in exchange for a review, ask only happy customers to review, swap reviews with other sellers, or pay a service for them. Amazon’s review policies prohibit all of it, and enforcement can include listing and account action. We go deeper in how to get your first Amazon reviews without risking your account.

Step 4: Build real sales with PPC

Sponsored Products ads are the most controllable lever a new product has. They put you on page one for a keyword long before you’ve earned it organically, and the sales they generate add to your product’s history.

How much do ad-driven sales help organic rank? Here we have to be honest. It’s widely believed in the industry that they do, and our experience points the same way, but Amazon hasn’t published how it treats them. So treat PPC as a way to generate real sales and data, not as a rank-buying machine.

A few launch habits that tend to work:

• Run exact match on your ladder keywords, so you can see which terms convert and how your ACoS (advertising cost of sales: ad spend divided by ad revenue) behaves on each.
• Run broad or auto campaigns alongside to discover terms you didn’t think of, then move winners into exact match.
• Expect higher ACoS early. A launch campaign is partly buying data. Set a budget you can live with for the first few weeks and decide in advance what “working” looks like.
• Watch conversion, not just clicks. If a keyword gets clicks but no sales, the listing (or the match) is the problem.

New sellers should also check Amazon’s US incentives page before starting. At the time of writing, it lists Sponsored Products credits of $50, $200 or $1,000 after matching spend, for a campaign created within 90 days of listing. Campaign structure is a big topic, so we’ve given it its own guide: Amazon PPC for a launch.

Step 5: Use discounts that live on Amazon

Price promotions can lift conversion at the moment it matters most. The key is that the discount must be part of the Amazon offer itself, visible to any customer.

Your main options:

• Coupons: a clippable discount shown in search. Amazon changed coupon fees from June 2, 2025, to $5 per coupon plus 2.5% of coupon-attributed sales, according to industry coverage of Amazon’s notice. New sellers may also qualify for a $50 coupon credit if they create a coupon within 90 days of listing.
• Prime Exclusive Discounts: a price cut only Prime members see, with a strikethrough on eligible listings.
• Lightning Deals and Best Deals: time-limited deals with their own fee structure ($70 per day plus 1% of deal sales, capped at $2,000, for non-peak deals after the June 2025 change, as reported at the time). Eligibility often depends on sales history and ratings, so many brand-new products can’t use them right away.

Fees and eligibility change often, so always confirm current terms in Seller Central before you plan around them.

The “gray area” that isn’t: big off-Amazon giveaways

Here’s where we have to be direct, because this is the tactic most new sellers are pitched. In November 2021, Amazon updated its Seller Code of Conduct to explicitly prohibit search rank manipulation. Its Seller Forums announcement named practices like two-step URLs, super URLs, funnels, treasure hunts and search-find-buy, along with off-Amazon rebates that make discounted sales look like full-price, organic ones. It also said sellers are responsible for what their service providers do.

So a service that hands out steep discounts or rebates off-Amazon and asks buyers to search a keyword and buy isn’t a gray area under that wording. It’s a stated violation. Some sellers still use versions of it and say they get away with it. We don’t recommend it. The upside is a few temporary positions. The downside can include your listing or your account.

Step 6: Add external traffic, but use it honestly

Traffic from outside Amazon (social, email, creators, your own website) can bring buyers who wouldn’t have found you through search. Brand Registered sellers can track it with Amazon Attribution, and the Brand Referral Bonus credits back a share of the referral fee on qualifying sales from non-Amazon traffic (industry sources describe an average of around 10%, varying by category).

Will it help organic rank? Possibly, if those visitors convert well, though Amazon hasn’t confirmed how external sales are weighted. Send people straight to your normal product page, at the normal price or an Amazon-visible promotion. That’s the line between legitimate external traffic and the funnels Amazon prohibits.

If your audience already lives on TikTok, that’s a natural source. Just remember the traffic quality matters more than the volume: a thousand curious clicks that don’t buy can drag conversion down.

Step 7: Protect what you’ve built

Ranks are earned slowly and lost quickly. A few things we keep seeing knock new products back down:

• Stockouts. If you go out of stock, you stop generating sales. Most sellers find rank falls sharply, and climbing back takes time and money. Plan inventory for success, not just for the forecast you hope to beat.
• Pulling ads too early. Turning PPC off the day you reach page one often means watching sales and position slide. Taper based on data instead.
• Price swings. A big jump right after a promotion ends can tank conversion.
• Ignoring early reviews. If the first five reviews mention the same flaw, that’s product feedback. Fix it before it becomes your star rating.

What’s genuinely hard (and what we’re unsure about)

We’d be overselling if we said a perfect checklist guarantees page one. Some things are just hard:

• Some categories are saturated. If the top results have tens of thousands of reviews and aggressive ad budgets, the head term may not be realistic in year one. Owning a cluster of specific terms can still be a good business.
• Budgets run out. A launch that needs months of heavy PPC to stay visible may have a product, price or margin problem rather than a ranking problem.
• The algorithm isn’t static. Amazon’s 2026 paper is a reminder that cold start handling has changed recently, and we can’t tell you exactly how much a launch today benefits compared with two years ago. Anyone who says they can is overconfident, ourselves included.

One lesson we’ve learned the hard way: when a launch stalls, the instinct is to add more of something (more ad spend, more coupons, more traffic). Usually the better move is to stop and ask which single step is broken. Is the product not indexed? Indexed but not clicked? Clicked but not bought? Each has a different fix.

If you want everything lined up before your next launch day, our Amazon product launch checklist covers the prep work.

Final thoughts

Learning how to rank a new product on Amazon is mostly about sequence. Make the listing unmistakably relevant. Pick keywords you can win now. Get honest reviews through approved channels. Use PPC and on-Amazon promotions to generate real sales. Then protect the position with inventory and steady pricing. Page ten is where almost everyone starts. It doesn’t have to be where you stay.

So, where is your launch stuck right now: getting seen, getting clicked or getting bought?

Related reading: Hiring an Amazon Product Launch Service: What’s Worth Paying For

If you’re stuck on this right now, the AZ Rank team can take a look. We’ve managed 6,000+ launches, and we’re a Helium 10 partner and Amazon SPN member. Reach us at azrank.com or hello@azrank.com.