Marketplace SEO

Marketplaces are search engines that happen to host products. Amazon’s A9 (and its successors), Etsy’s taste graph, eBay’s Cassini, and Walmart Marketplace’s ranking system each weight different signals — and a tactic that wins on Amazon will lose on Etsy. The cross-platform reality is brutal: $7 trillion of global commerce flows through marketplaces, and most retailers treat marketplace SEO as an afterthought.

TL;DR

Marketplace algorithms reward conversion velocity, not content quality. Amazon’s A9, eBay’s Cassini, and Walmart’s algorithm all heavily weight click-through, conversion rate, and sales velocity within the first 14-30 days of a listing. Etsy weights personalization on top of relevance.
Listing structure is platform-specific and unforgiving. Amazon’s title is 200 chars; Etsy’s is 140; eBay rewards 80-character titles; Walmart caps at 200. Each platform has hard rules that disqualify listings silently.
Off-marketplace signals matter increasingly. Amazon’s brand authority weighs external traffic; Etsy’s taste graph integrates Pinterest and Instagram signals; Google Shopping pulls from Amazon and Walmart for AI Overview citations.

The mental model

A marketplace algorithm is like a bouncer at a nightclub who keeps score by the second. They don’t care about your reputation across town; they care whether the people they let through the door are spending money inside. Stand in line longer? You move down. Get more turn-aways? You move down. Bring spending customers in fast? You move up — and once you’re a regular VIP, the line forms for you automatically.

That conversion velocity loop is the dominant signal across every major marketplace:

Listing surfaces in search results.
Click-through rate decides if it ranks higher next time.
Conversion rate decides if it ranks even higher.
Sales velocity (units × velocity) compounds the rank.
Reviews and seller metrics close the feedback loop.

Off-marketplace SEO (Google) rewards content authority. Marketplace SEO rewards transactional efficiency. The skills overlap, but the signals don’t.

Deep dive: the 2026 reality

Amazon and the A9 successor

Amazon retired the original “A9” name internally years ago, and the current ranking system (sometimes called COSMO or successor names depending on the source) blends classic A9 signals with deep-learning models. The factors that consistently move rank, validated by Helium 10, Jungle Scout, and Pacvue studies through 2025:

Sales velocity in the first 14, 30, and 90 days. New listings without sales never rank, regardless of optimization.
Conversion rate at the keyword level. Amazon tracks which queries lead to clicks on your listing and which clicks convert.
Title relevance — exact match for the primary keyword in the first 80 characters.
Bullet points and A+ Content as secondary keyword targets.
Backend search terms (250 bytes total) for synonyms and long-tail variants.
Product reviews — quantity, average, and recency. Recency weights heavier than total count.
Buy Box ownership — losing the Buy Box (lower price, faster fulfillment, FBA, etc.) drops rank.
PPC sales count toward organic ranking velocity. Amazon does not separate organic from paid sales for ranking purposes.
External traffic routed via Amazon Attribution links signals brand authority and lifts organic rank.

Amazon listing structure in 2026:

Title:        ≤200 characters total, primary keyword in first 80
Bullet 1-5:   ≤500 chars each (some categories 250). Lead with benefit, support with feature
Description:  ≤2,000 chars or A+ Content (Brand Registry required)
Backend:      Search Terms (250 bytes), Subject Matter, Other Attributes
Images:       Main image white background, 1000×1000 minimum (zoom enabled at 1600×1600+)
A+ Content:   Brand-registered sellers, 7 modules, ~5x conversion lift on tested SKUs
Brand Story:  Header carousel above A+, links across catalog

Amazon policy disqualifications that hide your listing or suppress it:

Promotional language in title or images (“Best seller!”, “Free shipping!”, “100% guaranteed!”).
Multiple products in main image (must be single product on white).
HTML in description (banned since 2023).
Misleading category placement.
Trademark infringement claims (handled in 24-72 hours; can suppress before resolution).
Variation policy abuse (treating different products as variants).

Etsy and the taste graph

Etsy’s algorithm has shifted significantly since 2019 toward personalized search (the taste graph). The signals that matter:

Listing quality score — clicks, favorites, conversions on the listing.
Recency — newly listed and recently renewed listings get a temporary boost.
Tag relevance — 13 tags max, 20 chars each. Each tag matches against query terms in order.
Title — first 40 chars carry highest weight, full 140 chars indexed.
Translations — Etsy auto-translates listings; native-language tags outperform.
Shipping costs — free shipping (or sub-$6 US shipping) gets a soft ranking boost.
Shop quality — shipping speed, message response time, review average.
Personalization — buyer’s history with similar styles, materials, price points reorders the SERP per user.
Off-site ads — Etsy’s Offsite Ads (off by default for sellers under $10K) feed back into ranking through external clicks.

Etsy’s taste graph integrates Pinterest, Instagram, and external pin/share signals. A listing that gets pinned on Pinterest sees Etsy traffic lift independent of Etsy’s internal search.

Etsy title structure that performs:

Bad:  Beautiful Handmade Necklace Gift for Her Love Birthday Present Special
Good: Birthstone Necklace, Gold Filled, Personalized Initial Pendant, Bridesmaid Gift

The good version leads with the search term, includes specific attributes (material, personalization), and names the gifting occasion.

eBay and Cassini

eBay’s Cassini ranking system has been in production since 2013 with periodic updates. Signals:

Click-through rate from search results.
Sales conversion rate at the listing level.
Sell-through rate for listings (how quickly inventory moves).
Listing freshness — Good ‘Til Cancelled listings need updates to stay competitive.
Title relevance — 80 characters max, every keyword counts.
Item specifics — brand, model, MPN, size, color, condition. Increasingly required by category.
Shipping cost and speed — free or cheap fast shipping ranks higher.
Return policy — 30-day or longer returns with free shipping rank higher.
Seller performance — Top Rated Seller status weights ranking visibly.

eBay title best practice: front-load keywords, use the full 80 chars, no promotional fluff (eBay won’t suppress like Amazon does, but Cassini deprioritizes).

Walmart Marketplace

Walmart’s algorithm is the youngest of the major marketplaces and the most opaque. Documented signals from Pacvue and Jungle Scout Walmart studies:

Listing Quality Score in Seller Center (out of 100).
Conversion rate (weighted heavily).
Buy Box ownership.
Pro Seller Badge — earned via on-time delivery, customer service, and policy compliance.
Walmart Fulfillment Services (WFS) parallels Amazon FBA and ranks higher than seller-fulfilled.
Sponsored Products budgets correlate with organic rank.

Walmart listing structure:

Product Name:    ≤200 chars, format: Brand + Defining Quality + Item + Style + Pack Count
Key Features:    3-10 bullets, ≤80 chars each
Description:     150-300 words, plain text (no HTML)
Images:          White background, 2000×2000+, 4-8 images recommended
Attributes:      Category-specific, 30+ fields per category

Cross-platform signals that move all marketplaces

Some signals matter everywhere:

Signal	Amazon	Etsy	eBay	Walmart
Conversion rate	Critical	Critical	Critical	Critical
Sales velocity	Critical	High	High	High
Reviews quantity	Critical	High	High	High
Reviews recency	High	Medium	Medium	High
Image quality	Critical	Critical	High	Critical
Free/cheap shipping	Critical (Prime)	Medium	High	High (WFS)
External traffic	Medium-High	Medium	Low	Medium
Seller metrics	High	High	High	Critical

The cross-platform tactics that work everywhere:

Lead with the searched-for term in the first 60-80 characters of the title.
White-background main image at the highest resolution allowed.
Optimize for conversion rate before keyword volume — a high-converting niche wins over a low-converting head term.
Drive external traffic to listings (Pinterest, Instagram, TikTok, brand site) — every marketplace rewards it, even if their docs are coy.
Maintain seller metrics ruthlessly; one cancelled order can hurt rank for weeks.

Visualizing it

flowchart TD
  A[Listing goes live] --> B[Indexed by marketplace]
  B --> C[Surfaces in SERP for relevant queries]
  C --> D{Click through rate}
  D -->|Above category median| E[Rank lifts]
  D -->|Below category median| F[Rank declines]
  E --> G{Conversion rate}
  F --> H[Buried beyond page 3]
  G -->|Above category median| I[Sales velocity grows]
  G -->|Below category median| J[Stalls or declines]
  I --> K[Reviews accumulate]
  K --> L[Reinforces rank]
  L --> C
  J --> M[Optimization cycle: title, images, price, reviews]
  M --> C

Bad vs. expert

The bad approach

A typical seller copies their Shopify PDP into Amazon Seller Central:

Title: Atelier Nord Red Leather Jacket - Premium Quality - Free Shipping!
Bullets:
- High quality
- Made with care
- You will love it
- Buy now
- Best deal ever

Description: Welcome to Atelier Nord. We are a small team of artisans...

No keyword research, no specs, promotional language (“Free Shipping!”), bullets that say nothing, brand-story description that doesn’t mention the product. Amazon’s category quality team will suppress the listing within 48 hours. Even if it survives, organic rank will plateau at zero — no conversions because no keywords match buyer intent.

The expert approach

Title: Atelier Nord Lambskin Moto Jacket for Women, Asymmetric Zip Leather Coat with Satin Lining, Italian-Tanned, Oxblood Red, Small

Bullets:
- ITALIAN LAMBSKIN: 100% genuine lambskin from Italian tanneries, softens with wear, breathable in 50-65F conditions, treated for water resistance
- ASYMMETRIC YKK ZIP: Nickel-free YKK Excella main zip, four functional pockets (two zippered chest, two snap waist), ideal for layering
- TRUE MOTO FIT: Tailored through the waist with snap collar, runs true to size for women, size up if layering over flannel or sweater
- TESTED PNW CONDITIONS: We field-tested across three Pacific Northwest winters, documented wear and care notes available on our brand store
- 60-DAY RETURNS: Free returns within 60 days, satin lining repairs covered first year, sizing exchanges expedited

Backend Search Terms (250 bytes):
oxblood leather jacket women, asymmetric moto jacket, italian leather coat, lambskin biker jacket, ladies leather jacket small, AN-RLM-001

Why it works: title leads with brand + material + product + critical attribute + variant; bullets lead with benefits in caps for scanability, support with concrete specs; backend search terms cover synonyms and the SKU. The listing answers the queries Amazon’s algorithm sees (“oxblood leather jacket women,” “lambskin moto jacket”) and converts visitors who land on it.

For Etsy, the same product needs different framing:

Title: Oxblood Red Leather Moto Jacket Women, Italian Lambskin, Asymmetric Zip, Handcrafted Small

Tags (13):
red leather jacket
moto jacket women
oxblood jacket
italian lambskin
asymmetric jacket
biker jacket women
leather coat small
handmade jacket
gift for her
fall jacket
small leather jacket
red biker jacket
custom moto jacket

Etsy weighs each tag as a discrete query target; Amazon weighs the title and backend terms as a single keyword pool. Same product, different optimization.

For eBay:

Title (80 chars): Atelier Nord Italian Lambskin Red Leather Moto Jacket Women's Small NWT 489

eBay’s 80-char limit forces ruthless prioritization. Note the inclusion of NWT (New With Tags) and price — both matched by buyer queries on eBay specifically.

For Walmart:

Product Name: Atelier Nord Women's Italian Lambskin Moto Jacket, Oxblood Red, Asymmetric Zip, Small, 1-Pack

Walmart prefers structured taxonomy in the name (Brand + Gender + Material + Item + Color + Style + Size + Pack Count).

Do this today

Rank-track on each marketplace separately. Use Helium 10 or Jungle Scout for Amazon; eRank or Marmalead for Etsy; Algopix or Title Builder for eBay; Pacvue or Helium 10’s Walmart tools for Walmart. Cross-marketplace tools don’t give you the same depth.
For Amazon: pull your top 10 SKUs by revenue. Open Brand Analytics → Search Terms Report (requires Brand Registry). Identify the top 50 search terms in your category. Map each term to the closest matching product. Any high-volume term you don’t rank for is a backend search terms or title optimization opportunity.
For Etsy: open Shop Manager → Stats → Search terms (Stats are free; eRank adds depth). Identify queries that bring traffic with low CTR — those listings need title or first-image work. Then audit your tags: every tag should be a real shopper phrase, not a single word.
For eBay: in Seller Hub → Performance → Traffic, sort listings by impressions vs sell-through rate. Listings with high impressions and low STR are either mispriced or photographed badly. Update title, photos, or item specifics — eBay rewards listing freshness within 30 days.
For Walmart: check Listing Quality Score in Seller Center. Anything under 90 has missing attributes, image issues, or content gaps. Each missing attribute costs ranking; fill every category-relevant field.
Cross-platform image audit: white-background main image, 1500-2000px on long edge, four to seven supplementary images per listing. Run through PageSpeed Insights to confirm load time on the marketplace’s PDP — slow listings convert worse, regardless of platform.
Conversion rate optimization beats keyword optimization on marketplaces. A/B test main images via Amazon’s Manage Your Experiments (Brand Registered sellers), Etsy’s listing photo position, eBay’s Promoted Listings impression test. The image change with the highest conversion lift wins regardless of “best practice” rules.
External traffic: build a Pinterest strategy with rich product pins linking to Etsy and Amazon. TikTok Shop integration drives both Amazon and brand-site traffic. Instagram shopping still moves needles for Etsy. Marketplaces measure external referral and reward it; the docs are vague but the data is consistent.
Set up automated review requests within marketplace TOS: Amazon’s “Request a Review” button via Seller Central or FeedbackWhiz; Etsy’s automated thank-you DMs; eBay’s feedback request flow. Review velocity within the first 14 days post-purchase moves rank measurably.
Build a brand presence off-marketplace that funnels back. Module 63’s Google Merchant Center work makes your own site competitive; the marketplace listings then benefit from external brand searches that lift trust signals across every platform you sell on.