ads-retail-media

Note: This is a leaf sub-skill. The thresholds and rules below live inline. In Wave 3, once

ads-walmart

ships and the

ads-retail-media

orchestrator extracts the shared retail-media logic, a dedicated

retail-media-specs.md

reference file will replace these inline rules (planned, not yet created — do not link to it from other files).

1. Collect Amazon Advertising export: Search Term Report (last 60 days), Campaign Performance Report, Targeting Report, Placement Report, Bulk Operations file
2. Collect Brand Analytics if available (Brand-Registered sellers / vendors): Search Terms, Repeat Purchase Behavior, Item Comparison
3. Verify Seller / Vendor Central context (vendor strategy differs from seller strategy on portfolio, coupons, and Sponsored Brands video)
4. Read

   ads/references/benchmarks.md

   for Amazon-specific benchmarks
5. Read

   ads/references/scoring-system.md

   for weighted scoring
6. Evaluate checks per category
7. Calculate Amazon Ads Health Score (0-100)
8. Generate findings report with action plan

• Portfolio organization — campaigns grouped by funnel stage (Awareness / Consideration / Conversion) or by product line, not ad-hoc
• Campaign naming convention consistent (e.g.,

  SP_Brand_Exact_USD_Conv

  /

  SB_NonBrand_HSA_Video_USD_Cons

  )
• Auto vs Manual mix: Sponsored Products has ≥1 Auto campaign per ASIN cluster for search-term harvesting (auto-mining workflow)
• ASIN-level coverage: every priority ASIN appears in ≥1 SP campaign (Exact, Phrase, or Broad) AND ≥1 Auto campaign
• Targeting type mix in SP: Manual Keyword + Manual Product (ASIN targeting) + Auto (for harvesting). Pure Auto is a learning-only state
• Sponsored Brands: at least one HSA (Headline Search Ad) campaign per brand storefront; SB Video tested for high-AOV products
• Sponsored Display: separate Audience (retargeting + lookalike) and Contextual (product / category targeting) campaigns — don't mix in one

• Auto → Manual harvest cadence weekly or bi-weekly: pull converting search terms from Auto, promote winners to Manual Exact in their target campaign, add as negatives to the Auto so spend redirects
• Negative keyword coverage: every Auto campaign has Manual campaign ASINs as negatives (prevents Auto cannibalizing Manual placements)
• Negative product targeting: irrelevant ASINs added at campaign level (prevents Sponsored Display contextual targeting from showing your ad next to incompatible products)
• Search Term Report harvested last 7 days — any high-spend, zero- conversion terms above $10 / 30 days should be negatives
• Brand defense: branded search terms are isolated in a separate campaign with exact-match brand keywords and competitor-name negatives
• Top-of-search visibility — branded search auction CPC vs default; flag if competitor bid is above your brand-defense bid

• ACOS targets set per portfolio based on contribution margin (target ACOS = (1 - contribution_margin) × buffer). Sloppy "30% across the board" ACOS targets cap performance
• TACOS (Total ACOS) trending down quarter-over-quarter — TACOS measures total ad spend / total revenue including organic; downward trend means paid is lifting organic, not replacing it
• ROAS reported alongside ACOS for the team that thinks in ROAS
• Coupon + Lightning Deal stacking — when on, ACOS targets relaxed to capture lift; auto-flag campaigns that need temporary ACOS uplift
• Day-parting: bid adjustments by hour for ASINs where add-to-cart hours differ from purchase hours (frequent for high-AOV electronics)

• Dynamic bidding strategy:

  Up and Down

  for Manual Exact converting campaigns;

  Down Only

  for Auto campaigns or Sponsored Display contextual (avoid bid escalation in learning campaigns)
• Placement bid multipliers: Top-of-search placement adjusted based on ROAS by placement (Placement Report); often +50-100% for top-of-search, -30% for product page placements
• Budget caps appropriate — no campaigns capped at <2x current daily spend (caps strangle algorithm learning)
• Bulk bid adjustments — bid optimization runs weekly via Bulk Operations not ad-hoc in the UI

• HSA (Headline Search Ad): brand logo + 3-product showcase + landing page (Store, custom URL, or product list)
• SB Video: 6-45s; 15-30s sweet spot. Hook within 2s. Captions on. Required for high-AOV products
• Brand Store linked as landing destination where available (higher CVR than product detail page)
• Negative keywords on SB include competitor brand names if you don't want top-of-funnel competitor traffic; remove if you do

• Audience targeting: Views Remarketing, Purchase Remarketing, Cross- Sell, and Lookalike audiences each scaled with appropriate bids
• Contextual targeting: relevant categories AND specific ASIN targeting (your own complementary products + competitor ASINs you can win)
• Display creative quality: at minimum, the default product creative; custom creative tested for top SKUs
• Off-Amazon Sponsored Display (Twitch, IMDb, third-party sites) audited separately — different CPC dynamics, different attribution windows

• Brand Analytics accessed (Brand-Registered only): Top Search Terms, Item Comparison, Repeat Purchase Behavior
• Top Search Terms report used to identify share-of-voice gaps vs competitors on category keywords
• Repeat Purchase Behavior informs subscription / replenishment ad strategy
• Amazon Attribution (off-Amazon traffic) tracked if running external ads driving to Amazon listings

• AMAZON-ADS-REPORT.md

  : Full per-category findings
• Search-term harvest plan: keywords to promote to Manual + keywords to add as negatives, ranked by spend recovery $
• ACOS-by-ASIN matrix highlighting unprofitable SKUs
• Sponsored Brands video opportunity list (top 10 ASINs without SB Video)
• Pre-launch checklist when entering a new product category
• Amazon DSP entry recommendation (yes / no based on total monthly spend and audience size)