Walmart Seller Analyzer | Product Samples, Scenarios & Niche Research

Analyze public Walmart US product observations, storefront samples or keyword search results. Get observed price, seller, reviews, badges, availability, fulfillment signals and explicitly assumed cost/opportunity scenarios. These are public samples, not verified seller turnover or private Seller Center sales. Unknown demand remains null. No Walmart login/API key required; Apify API/MCP requires your token/account.

First run

{"mode":"niche","searchQueries":["air fryer"],"maxResultsPerQuery":3,"maxDetailedProducts":3}

Use product rows for comparison; inspect SUMMARY catalogs for coverage/continuation and ERRORS for source failures. Export JSON/CSV/Excel through Apify. A product can be useful with unknown revenue; do not interpret null as zero sales.

Modes

• seller: sellerUrls such as https://www.walmart.com/global/seller/101644973. Returns observed storefront products plus a free seller_summary. Storefront carousels are incomplete and subsequent pages may repeat; catalogCoverageComplete remains false. This is not an entire seller catalog or turnover report. requestedSellerId/sellerAttribution distinguish the requested storefront from the featured product seller. Association does not prove exclusive seller sales or Buy Box share.
• product: productUrls or numeric productIds, e.g. 16954719851. Product identity must match the requested ID. Duplicate products across inputs are delivered once.
• niche: searchQueries, e.g. air fryer. Real keyword pages are read and deduplicated, then the collected sample is ordered by a heuristic opportunity score. This is not Walmart's official ranking or a full market census.

Interpret the evidence: agents and workflows

price, rating, reviewCount, sponsored, isWfs and offerCount remain null when unavailable. offerCount is an observed count when present; no complete secondary offers list or Buy Box-share estimate is promised. detailStatus identifies listing_only, complete, source_unavailable or identity_mismatch. Failed detail requests retain useful listing observations with warnings.

An observed bought-since-yesterday bucket is extrapolated into hypothetical monthly units as 30/42/60 times the bucket count, assuming it repeats. This is a scenario, not a measured monthly range, guaranteed lower bound or confidence interval. Lifetime reviews and cart counts do not produce monthly sales estimates. Without a usable purchase badge, estimatedMonthlyUnits/Revenue/Profit fields are null. Estimates concern the product across sellers; estimatedSellerRevenueMid remains null.

seller_summary exposes matched/sample/estimate counts and modeledProductRevenueMidSubtotal for products with usable signals only. Its estimatedMonthlyRevenueMid and estimatedSellerRevenueMid stay null: no entire seller revenue is established. estimateCoverageComplete is about modeled rows, separate from catalogCoverageComplete. Do not treat an incomplete subtotal as a complete portfolio.

Profit is a scenario using COGS, shipping, WFS, advertising, refunds and an assumed category referral fee. Optional referralFeePercent overrides the assumption for your contract. Defaults are COGS30%, refund2%, shipping/WFS/ads0. ROI uses COGS as its base; opportunityScore and dataCoverageScore are heuristics, not accuracy, probability or investment advice. Verify current contract-specific rates with the official fee schedule.

requestedPostalCode/requestedStoreId are recorded intent only. This Actor does not force or verify a session ZIP/store. locationText/observedPostalCode are observations; mismatch warnings must be retained when comparing prices and availability. Missing location is unknown, not confirmation of the requested ZIP.

Input limits

Field	Default	Meaning
mode	niche	seller, product or niche
sellerUrls/productUrls/productIds/searchQueries	mode-dependent	Maximum20 targets per array; public HTTPS Walmart URLs and numeric item/storefront IDs only
maxProductsPerSeller/maxResultsPerQuery	5	1–500 sample product limit per target; event budget may return fewer
includeProductDetails	true	Fetch optional exact-item details
maxDetailedProducts	5	0–100 detail attempts; zero disables enrichment
maxConcurrency	2	1–10 parallel detail requests
startPage/startOffset	1/0	Start page and unique-card offset on that page
maxPages	10	1–20 catalog requests; cap/failed/repeated pages produce partial coverage
postalCode/storeId	95829/empty	Recorded requested location only
costOfGoodsPercent/refundRatePercent/advertisingRatePercent	30/2/0	Cost scenario percentages
shippingCostPerOrderUsd/wfsFeePerOrderUsd	0/0	Per-order scenario costs
referralFeePercent	category assumption	Optional0–100 override
proxyConfiguration	Apify Residential	Supports Apify proxy settings or custom proxyUrls; proxy/compute costs are separate

Use SUMMARY catalog nextInput to continue, preserving your limits and cost assumptions. Ordering can change between requests; compare IDs. Seller repeated-page detection discloses source limits and does not fabricate new pages. Complete search coverage is claimed only after a valid empty page, not after reaching a row/page cap. Change monitoring is not implemented in this release.

Pricing and failures

Product analyzed: $2 per 1,000 useful delivered rows on every plan. Startup is $0.00005 per GB, minimum one event, not an unconditional per-run fee at any memory size. User also pays platform/proxy usage. Seller summaries are free. Failed/skipped products have no product event; source attempts can still consume resources. A useful product observation with unknown estimates is billable once.

Set maxTotalChargeUsd on the run API to cap event charges; startup is accounted for before product work. SUMMARY budgetLimited discloses stops. Errors are stored in ERRORS, not paid error rows. Pure source failure with no observations fails; a valid empty search can succeed. Public Walmart access can be temporarily unavailable or return a challenge. A successful run alone does not establish complete or accurate financial estimates.

REST API

curl -X POST 'https://api.apify.com/v2/acts/apivault_labs~walmart-seller-revenue-product-opportunity-analyzer/runs?build=latest&memory=512&timeout=360&maxTotalChargeUsd=0.05' \
  -H "Authorization: Bearer $APIFY_TOKEN" -H 'Content-Type: application/json' \
  -d '{"mode":"product","productIds":["16954719851"]}'

Poll /v2/actor-runs/{runId} to terminal. Read /v2/datasets/{defaultDatasetId}/items?clean=true (format=csv for export), then /v2/key-value-stores/{defaultKeyValueStoreId}/records/SUMMARY and ERRORS. Keep tokens out of shared input/URLs; use Bearer headers.

MCP and agent instructions

Connect an authenticated client to https://mcp.apify.com/?tools=apivault_labs/walmart-seller-revenue-product-opportunity-analyzer. Discover the Actor with tools/list and use the exact returned tool name. The hosted server currently abbreviates this long slug to apivault_labs--walmart-seller-revenue-product-opportunity-a-dbe5. tools/call takes normal inputs plus waitSecs:30. Use returned run/storage IDs to poll and read get-dataset-items and get-key-value-store-record for SUMMARY/ERRORS. For strict event caps start through REST maxTotalChargeUsd.

Agents must preserve nulls, sellerAttribution, estimateScope/status, detailStatus, observed location, warnings and catalog coverage. Keep source observations separate from extrapolated scenarios; never describe a product-wide estimate or sample subtotal as seller turnover. Follow actual nextInput and inspect stopReason rather than assuming complete pagination. In n8n/Make/Zapier map those evidence fields alongside price/ID, and split free seller_summary rows from paid products.