Marketplace Advertising

Marketplace Advertising

Overview

Marketplace advertising (Amazon Sponsored Products, Walmart Connect, eBay Promoted Listings) puts your products in front of shoppers with high purchase intent directly on the platform where they are shopping. Unlike Google/Meta, marketplace ads run within a closed ecosystem — clicks and conversions happen on the marketplace, not your site. The strategic work is campaign structure, keyword harvesting, and bid management. Amazon's native Campaign Manager UI handles most of this without code; automation tools like Perpetua or Sellics add bid optimization rules on top.

When to Use This Skill

• When selling on Amazon and needing a structured campaign setup from scratch
• When ACOS (Advertising Cost of Sale) is above your target and needs systematic optimization
• When running campaigns across multiple marketplaces and needing unified reporting
• When wanting to automate keyword harvesting and bid adjustments without building custom scripts
• When launching a new product and needing an auto-campaign to discover converting search terms

Core Instructions

Step 1: Choose the right advertising management approach

Start with Amazon's native Campaign Manager for campaign setup. Once you have 30 days of data, add Perpetua or Jungle Scout for automated bid optimization if managing more than 5 active campaigns.

Step 2: Set up your Amazon Sponsored Products campaigns

Note: Marketplace advertising is independent of your Shopify/WooCommerce/BigCommerce store. Amazon campaigns run within Seller Central regardless of your ecommerce platform.

---

Amazon Seller Central — Campaign Setup

Create the two-campaign structure for each product group:

Auto campaign (keyword discovery): 1. Go to Amazon Seller Central → Advertising → Campaign Manager → Create Campaign 2. Select Sponsored Products 3. Set:

• Campaign name: [Product Name] - Auto
• Daily budget: 30% of your total product budget (e.g., $15/day if total budget is $50/day)
• Targeting: Automatic targeting
• Default bid: $0.75

4. Under Automatic Targeting, set bids for all four match types (close match, loose match, substitutes, complements) — start equal, optimize after 2 weeks of data 5. Create the ad by selecting your ASIN

Manual campaign (scaling proven keywords): 1. Create a second campaign: [Product Name] - Manual 2. Set:

• Daily budget: 70% of your total product budget
• Targeting: Manual targeting → Keyword targeting
• Match types: Add each seed keyword in both Exact (bid: $1.20) and Phrase (bid: $0.90)

3. Add negative keywords at campaign level: "free", "DIY", "how to", competitor brand names

---

Harvest keywords from auto campaign (weekly task)

After 7–14 days, move converting search terms from auto to manual:

1. Go to Campaign Manager → [Auto Campaign] → Search Term Report 2. Download the report for the last 14 days 3. Filter for rows where:

• 7 Day Total Orders (#) > 0
• Advertising Cost of Sales (ACOS) < your target ACOS (e.g., 30%)
• Clicks >= 5 (minimum data threshold)

4. For each qualifying search term:

• Add it as an Exact match keyword in your manual campaign
• Add it as a Negative exact keyword in your auto campaign (prevents both campaigns from competing for the same query)

---

Walmart Campaign Center

1. Go to Walmart Seller Center → Advertising → Campaign Manager 2. Create a Sponsored Products campaign 3. Walmart's setup mirrors Amazon's: choose automatic or manual targeting, set a daily budget and default bid 4. Walmart's search term report works identically to Amazon's — run the same weekly harvest process

---

eBay Promoted Listings

1. Go to eBay Seller Hub → Marketing → Promoted Listings 2. Select Promoted Listings Standard (cost-per-sale, no cost for clicks) 3. Set an ad rate (typically 5–15% of item sale price — higher rates = more visibility) 4. Select listings to promote — focus on listings with high conversion rates 5. eBay does not require keyword management for Standard; use Promoted Listings Advanced (CPC model) for keyword-level control

Step 3: Set ACOS targets by product margin

ACOS = Advertising Cost ÷ Attributed Sales. Your target ACOS depends directly on your margin:

Set your target ACOS in Campaign Manager → Bidding Strategy → Target ACOS (dynamic bidding) — Amazon will automatically adjust bids up/down to hit your target.

Step 4: Bid optimization

Amazon's built-in dynamic bidding (no third-party tool needed): 1. Go to Campaign Manager → [Campaign] → Edit 2. Under Bidding, select Dynamic bids — up and down 3. Set a target ACOS — Amazon adjusts bids in real time 4. This works well for established campaigns with 30+ days of data

Manual bid optimization rules (do weekly):

• Keywords with ACOS > 1.2× target and 10+ clicks → reduce bid by 15%
• Keywords with ACOS < 0.8× target and 1,000+ impressions → increase bid by 10%
• Keywords with 20+ clicks and zero sales → pause or add as negative

With Perpetua (automated): 1. Connect Perpetua to your Amazon Seller Central account 2. Set a goal (target ACOS or target revenue) per campaign 3. Perpetua runs bid optimization daily using machine learning — no manual weekly reviews needed 4. Go to Perpetua → Campaigns → Performance to review automated recommendations before they execute

Step 5: Track performance

Monitor these metrics in Campaign Manager weekly:

Step 6: Custom / API-based management

Use the Amazon Advertising API only if you are managing 50+ campaigns programmatically or building a custom reporting dashboard. For most sellers, Campaign Manager is sufficient.

// Amazon Advertising API client — only needed for programmatic campaign management
async function getAmazonAdsToken(config: {
  clientId: string;
  clientSecret: string;
  refreshToken: string;
}): Promise<string> {
  const response = await fetch('https://api.amazon.com/auth/o2/token', {
    method: 'POST',
    headers: { 'Content-Type': 'application/x-www-form-urlencoded' },
    body: new URLSearchParams({
      grant_type: 'refresh_token',
      client_id: config.clientId,
      client_secret: config.clientSecret,
      refresh_token: config.refreshToken,
    }),
  });
  const { access_token } = await response.json();
  return access_token;
}

// Automated bid optimization — runs weekly as a scheduled job
async function optimizeBids(adGroupId: string, targetAcos: number) {
  // Get 30-day keyword performance from the Reporting API
  // Filter keywords with insufficient data (< 10 clicks) — skip them
  // Keywords with ACOS > targetAcos * 1.2: reduce bid 15%
  // Keywords with ACOS < targetAcos * 0.8 and impressions > 1000: increase bid 10%
  // Clamp all bids between $0.10 and $10.00
}

For most sellers, use Perpetua or Jungle Scout's built-in automation rather than the API — these tools provide the same bid logic with a UI and no infrastructure to manage.

Best Practices

• Separate auto and manual campaigns from day one — auto campaigns discover keywords; manual campaigns scale the winners with controlled bids; mixing them creates reporting confusion
• Harvest keywords weekly for the first 60 days — after initial launch, graduate the best terms to manual and add non-converters as negatives in the auto campaign
• Pause keywords spending $20+ with zero sales — set this as a weekly rule; low-converting keywords drain budget without contributing to ACOS
• Use campaign-level negatives before launch — add "free", "DIY", "how to", "used", "wholesale" as broad match negatives to all campaigns at launch
• Align ACOS targets with actual product margins — a 30% ACOS target on a 25% margin product guarantees losses; recalculate for each product group
• Run Sponsored Brands alongside Sponsored Products — occupying both the top banner (Sponsored Brands) and product grid (Sponsored Products) increases share of voice significantly

Common Pitfalls

Related Skills

• @google-shopping-feed
• @google-ads-ecommerce
• @marketing-attribution-dashboard
• @multi-channel-selling
• @pricing-promotions

Automated Keyword Bid Optimizer

Problem/Feature Description

An Amazon seller managing dozens of Sponsored Products ad groups is spending hours each week manually reviewing keyword performance and adjusting bids in the advertising console. Their ACOS is inconsistent across the catalog — some keywords are underperforming and draining budget, while others are highly efficient but stuck at low bids that limit their visibility. The seller wants to automate this weekly bid review process.

The optimization logic needs to be data-driven: keywords without enough performance history shouldn't be touched, and bid changes should only be applied when there's a meaningful difference to avoid unnecessary API churn. The system must also handle the reality that Amazon's API has rate limits and that making the same change twice before it propagates causes confusing state. The target ACOS should be configurable per product line.

Output Specification

Write a TypeScript file named bid-optimizer.ts that:

• Implements a function to optimize bids for all enabled keywords in a given ad group based on historical performance
• Applies directional bid adjustments based on ACOS relative to the target
• Respects minimum data requirements before making changes
• Enforces bid floor and ceiling bounds
• Only submits API updates when the bid change is meaningful
• Handles Amazon API reliability concerns appropriately

The implementation should use fetch for HTTP calls and follow the Amazon Ads API authentication pattern. Include inline comments explaining the logic for each decision threshold.

Amazon Sponsored Products Campaign Setup

Problem/Feature Description

A small home goods brand just launched a new stainless steel travel mug on Amazon (ASIN: B0EXAMPLE123) and needs to start advertising. The brand manager has heard that a structured approach to Sponsored Products works better than a single campaign, but doesn't know how to set it up programmatically. They have the Amazon Advertising API credentials already registered in the console, and want a TypeScript module that creates the complete initial campaign structure for the product.

The brand operates in the North America marketplace, has a daily advertising budget of $50, and has identified initial seed keywords: "stainless steel travel mug", "insulated coffee tumbler", and "leak proof travel cup". They want the setup to be production-ready with proper authentication handling and structured so that it can easily discover new keywords over time while maintaining controlled spending on proven terms.

Output Specification

Write a TypeScript file named campaign-setup.ts that:

• Defines the Amazon Ads API configuration interface and authentication
• Implements a function to create the full campaign structure for the ASIN above
• Creates all necessary campaigns, ad groups, product ads, and keywords
• Is ready to be called with the provided parameters

The file should be self-contained and use only standard Node.js/TypeScript patterns with fetch for HTTP calls. No additional libraries beyond date-fns for date formatting are needed.

Sponsored Products Keyword Harvesting Tool

Problem/Feature Description

An Amazon seller running Sponsored Products campaigns has noticed their auto campaign is finding converting search terms, but those terms are not being captured in their manual campaign for more precise bidding control. They need a scheduled harvesting process that automatically identifies high-performing search queries from their auto campaign and promotes them to their manual campaign so they can bid more efficiently on proven terms.

The seller's marketing operations team wants this as a TypeScript module they can run as a scheduled automation job. The system should be smart about which queries to promote — not every query that gets a click is worth promoting, and they don't want to flood the manual campaign with untested terms. Once a term is promoted, the auto campaign should stop competing on it to avoid bidding against themselves.

The seller uses a target ACOS that varies by product, so the threshold should be configurable via environment variable rather than hardcoded.

Output Specification

Write a TypeScript file named keyword-harvester.ts that:

• Implements a function to harvest converting keywords from an auto campaign and add them to a manual ad group
• Pulls the appropriate search term report with correct parameters
• Applies filtering logic to select only statistically reliable, high-performing queries
• Adds qualifying queries to the manual campaign and prevents the auto campaign from competing on them
• Reads the target ACOS from an environment variable (fall back to a sensible default if not set)

The implementation should use fetch for HTTP calls and follow the same Amazon Ads API authentication pattern (config object with clientId, clientSecret, refreshToken, profileId, region).