1688 Pet Supplies Scraper — Wholesale Sourcing & Dropship Data

Research winning pet products on 1688.com — Alibaba's domestic wholesale marketplace — with the data you need to evaluate them, not just a Chinese title and a price you can't interpret.

Most 1688 scrapers dump Chinese-only titles and a single "from" price that hides the quantity breaks, the MOQ, and the supplier trust signals you actually need. This 1688 scraper takes an English keyword like "dog leash" and returns a research-ready row: an AI-translated English title, the full quantity-break price ladder, a 0–100 dropship-ready score, an estimated landed cost in USD, and compliance hints for your target market — all of these for the products covered by the per-run detail budget (see the SUMMARY diagnostic section below for what happens beyond it).

No Chinese needed. No 1688 account. No manual translation. One API call.

Keywords: 1688, 1688 scraper, wholesale, dropshipping, supplier search, product research, MOQ, landed cost

---

🛡️ Trust at a glance

What we commit to	How it works here
Pay only for delivered results	Zero-result runs cost $0.00 — you are billed per product row actually delivered, nothing else
No hidden usage fees	Pure pay-per-event pricing; proxy and platform compute costs are included, never passed through
Honest degradation	Every run's SUMMARY discloses exactly what happened — partial deliveries always come with a named reason, never silently
Flat fair pricing	One honest price per delivered row ($0.0045 + $0.01 run fee) — no tier tables, no subscription, no start fee
MCP / agentic-payment ready	Input, dataset, key-value-store and output schemas are all declared — machines can validate the contract before calling
Issues answered fast	Open an Issue instead of guessing — our response time is public on the developer profile
Live reliability stats	Success rate and run volume are shown live on this Store page — judge from real data, not claims

---

⚡ One-click examples

Skip the input form — these public tasks run proven configurations in one click:

• Quick Start — 10 products — the cheapest way to see a full research row (~$0.06)
• Deep Research — 50 full-detail items — two keywords, 50 detailed products with price ladders and landed cost
• Price Ladder Scan — cheapest first — results sorted by price for instant comparison
• Offer ID Lookup — full detail for known 1688 product IDs, no keyword search
• Fast Catalog Scan — 30 items in seconds — search-card-level rows, fastest and cheapest scan mode

---

🆚 Why 1688 instead of Alibaba.com or AliExpress?

1688.com is the Chinese domestic wholesale arm of the Alibaba group — the same factories that list on Alibaba.com sell there for the home market, typically at noticeably lower factory prices because there is no export markup. That makes 1688 the sharper place to do product selection and price comparison before you commit to a supplier. This Actor covers exactly that research step: search, compare and cost out products on 1688 from English. To be clear about scope: this Actor scrapes 1688.com (China domestic B2B), not Alibaba.com international — if you are weighing 1688 vs Alibaba as a sourcing channel or looking for an Alibaba alternative for price research, this is the tool that reads the domestic side.

---

✅ What you get / ❌ What this isn't

✅ What you get	❌ What this isn't
English titles (AI-translated) alongside the Chinese original	Not a Chinese-only export you paste into Google Translate column by column
Dropship-Ready Score (0–100) based on 7 weighted factors	Not a raw data dump that leaves you guessing if a product is worth researching
Estimated landed cost in USD (product + shipping + duty + fees)	Not a CNY price that ignores your real cost to land it in your market
Compliance hints (FDA / CE / Prop 65 / FCC) per target market	Not a scraper that leaves compliance research entirely to you
Pet-supplies keyword mapping (500+ EN→CN terms built in)	Not a generic scraper where "dog leash" returns zero Chinese results
Dropship-only filter — verified dropship products kept, proven non-dropship culled, unverifiable rows delivered but clearly flagged (dropshipStatus: "unknown")	Not an unfiltered list where half the results don't support dropshipping
PPE pricing $0.0045 per delivered product + $0.01 run fee, no startup fee (no monthly fee)	Not a $20–30/month subscription you pay even when you don't run it

---

🎯 Who is this for?

• 🚀 Dropshipping beginners testing pet products with low MOQ
• 📦 Amazon FBA sellers researching pet supplies from China factories
• 🐾 Pet niche store owners on Shopify looking for trending products
• 💰 Product researchers who need landed cost + margin before committing
• 🌐 E-commerce entrepreneurs who don't read Chinese but research products from China

---

🔧 Features

🌐 AI-Translated English Titles Every product comes with an AI-translated English title alongside the original Chinese title. No more pasting Chinese text into Google Translate column by column — you see a clean English title right away.

📦 Dropship-Ready Score (0–100) A weighted score based on 7 factors that tell you if a product is actually suitable for dropshipping:

• Dropship support (does the supplier offer one-piece dropship?)
• MOQ (lower is better for dropshipping)
• 48-hour pickup rate (how fast does the supplier ship?)
• Distributor count (how many dropshippers already work with them?)
• Supplier composite score (trust signal)
• 30-day order volume (demand signal)
• Repurchase rate (quality signal)

The score maps to a letter grade (A–D) so you can filter at a glance.

💰 Estimated Landed Cost in USD See your true per-unit cost — not just a CNY wholesale price. The estimate includes:

• Product unit cost (converted from CNY at a fixed FX rate)
• International shipping (estimated by weight and destination)
• Import duty (based on target market and product category)
• Platform fees

⚠️ Compliance Hints Get alerts about certification requirements for your target market before you list:

• US: FDA, FCC, CPSC, Prop 65, ASTM F963
• EU: CE, REACH
• UK: UKCA
• AU: ACCC

Compliance hints are heuristic estimates based on product titles and categories — always verify with a qualified expert.

🐾 Pet-Supplies Keyword Mapping 500+ English-to-Chinese keyword mappings built specifically for pet supplies. Type "dog leash" and the Actor automatically searches using the correct Chinese keyword. No need to know Chinese or use a separate translation tool.

🎯 Dropship-Only Filter Turn on dropshipOnly to exclude any product that doesn't explicitly support one-piece dropshipping. No more wading through wholesale-only listings that require bulk orders.

One expectation to manage: when a product's details cannot be fetched (detail requests are the platform's most heavily protected layer), items whose dropship status could not be verified are still returned and stamped dropshipStatus: "unknown" — they are not silently dropped. On large batches (>50 products) expect a meaningful share of items in this unverified state. Every run's SUMMARY reports the exact arithmetic: dropshipCulled (removed because the details proved they do NOT support dropship) and dropshipUnknownDelivered (returned unverified). You can filter strictly on your side by dropping rows where dropshipStatus = "unknown" — SUMMARY's dropshipCulled / dropshipUnknownDelivered counts let you verify the exact split.

📊 Quantity-Break Price Ladder Get the full price tier breakdown — not just a single "from" price. See exactly how much you pay at 1 unit, 100 units, 500 units, so you can plan your margins at scale. (Delivered for products covered by the detail budget — see the SUMMARY diagnostic section below.)

---

Quick Start

1. Run on Apify Console

1. Go to the Actor page on Apify Store.
2. Click Try for free.
3. Enter keywords like ["dog leash", "cat toy"].
4. Click Start and wait for results.
5. Download the dataset as JSON, CSV, or Excel.

2. Run via API

curl -X POST \
  "https://api.apify.com/v2/acts/china-pet-supplies-scraper/runs?token=YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "keywords": ["dog leash", "cat toy"],
    "maxResults": 50,
    "dropshipOnly": true,
    "targetMarket": "US"
  }'

Time budget for bigger orders: orders with maxResults above 55 automatically switch to an extended time budget (780 seconds) to fit the larger workload — nothing to configure. Orders of 55 items or fewer per keyword run within the standard 330-second budget, which may not be enough when you request several keywords at once; in that case the Actor stops at the deadline, marks the run partial in the summary, and you are billed only for the items actually delivered.

3. Run locally

npm install
apify login
apify run --input '{"keywords":["dog leash"]}'

---

🤖 Copy to your AI assistant

This Actor is API-first and agent-ready: deterministic JSON input/output, per-event billing with a documented floor, and machine-readable failure semantics (SUMMARY diagnostics) — ideal for MCP servers, n8n/Make flows, LangChain tools and autonomous sourcing agents.

Paste this into your AI assistant:

Use the Apify Actor crawleast/china-pet-supplies-scraper to research
wholesale pet products on 1688.com. POST this JSON to
https://api.apify.com/v2/acts/crawleast~china-pet-supplies-scraper/run-sync-get-dataset-items?token=<APIFY_TOKEN>&timeout=900
body: {"keywords": ["dog leash"], "maxResults": 20, "dropshipOnly": true,
"targetMarket": "US"}.
Each dataset row is one product with English title, quantity-break price
ladder, dropship-ready score and estimated landed cost in USD. Report the
top 5 by dropshipReadyScore, and flag any compliance hints.

run-sync-get-dataset-items returns the final dataset in one call — perfect for tool-calling loops. Larger orders (more than ~55 items, or several keywords at once) can take up to 15 minutes: append ?timeout=900 (seconds) to the sync endpoint, or set your HTTP client timeout ≥ 15 minutes — otherwise the gateway may return 408 while the run keeps working in the background (poll runs/{runId} to recover; delivered rows are billed per row regardless).

---

📜 Release notes: version history and upgrade guidance (0.3.36 → 0.4.2) live in ./CHANGELOG.md.

---

Input Parameters

Parameter	Type	Required	Default	Description
keywords	array<string>	✅	—	English search keywords, e.g. ["dog leash", "cat toy"]. API callers may also send query or search (string or array) as an alias when keywords is absent.
petCategory	enum	❌	all	clothing_leash / bed_cage / cleaning_grooming / toys / training / aquarium / all
maxResults	integer	❌	50	Max products per keyword (1–2000)
fullDetailSharding	boolean	❌	auto	Full-detail sharding for large orders: auto-ON when maxResults > 55, auto-OFF at ≤ 55. true forces sharding; false restores the legacy mixed shape (card-level surplus). See "Full-detail sharding" below.
maxRunTimeSecs	integer	❌	780	Runtime budget in seconds (max 780). The effective budget is the minimum of this value, the platform run timeout minus a 48s finish reserve, and the code-side cap — 330s for non-sharded runs (unchanged behaviour) and 780s when full-detail sharding is active. ⚠️ The 780s budget requires the platform run timeout of 900s (15 min) configured in actor.json — if you launch runs with a custom timeout below 900s, large orders get killed mid-shard (see "Full-detail sharding").
priceMinCNY	integer	❌	—	Min price in CNY — an APPROXIMATE search-level filter (see note below the table)
priceMaxCNY	integer	❌	—	Max price in CNY — an APPROXIMATE search-level filter (see note below the table)
dropshipOnly	boolean	❌	true	Only return products that support dropshipping. Honest caveat: items whose dropship support could not be verified (unknown state, e.g. search-card rows whose details could not be fetched) are NOT silently excluded — they are delivered and stamped dropshipStatus: "unknown" so you can tell verified rows from unverified ones. This fail-open is deliberate: strict exclusion of unknown-state items would jeopardize filling large orders.
maxMOQ	integer	❌	—	Max minimum order quantity (1–10000). Needs fetched detail data — in a degraded run, rows without details pass through (see note below).
merchantType	enum	❌	any	any / superFactory (Source Factory) / verifiedMerchant (Verified Merchant)
province	string	❌	—	Supplier province (Chinese), e.g. "浙江". Needs fetched detail data — in a degraded run, rows without details pass through (see note below).
city	string	❌	—	Supplier city (Chinese), e.g. "温州市"
sortBy	enum	❌	relevance	relevance / bestSelling / priceAsc / priceDesc / dropshipScore. priceAsc / priceDesc / dropshipScore re-order the ENTIRE delivered dataset before hand-off (items with no price sink to the end of a price sort; dropshipScore = highest dropship-readiness score first). Note: these three sorting modes hold all results back and deliver them in ONE ordered batch when the run finishes — the dataset is empty until then (expected); relevance / bestSelling deliver results as they are collected.
includeEnglishTranslation	boolean	❌	true	Translate titles to English
includeLandedCost	boolean	❌	true	Include landed cost estimate in USD
includeCompliance	boolean	❌	true	Include compliance hints for target market
targetMarket	enum	❌	US	US / EU / UK / AU / global
shippingWeightMax	number	❌	—	Max product weight (kg)
excludeFoodProducts	boolean	❌	true	Exclude pet food/treats (complex import requirements)
includeSkuDetails	boolean	❌	false	Fetch detailed SKU specs (slower)
includeDescriptionImages	boolean	❌	true	Fetch product description images (adds 2–3s per product)
offerIds	array<string>	❌	[]	Direct product IDs to fetch, bypassing keyword search
maxPages	integer	❌	—	Max search pages per keyword (1–100)
skipDetails	boolean	❌	false	Skip the detail layer entirely for fast, cheap search-card-level runs
proxyConfiguration	proxy	❌	—	Apify proxy config (use RESIDENTIAL CN for production)

How the filters really work (transparency note):

• Price filters are approximate. priceMinCNY / priceMaxCNY are applied on the SEARCH layer against the listing price shown on each search card. The product page may use tiered (quantity) pricing that differs from the card price, so an occasional item slightly outside your range can still be delivered — treat the bounds as a strong narrowing, not a hard guarantee.
• Detail-dependent filters fail-open in degraded runs. Filters that need fetched product details (maxMOQ, province, city, and similar) can only check items that reached the detail layer. If a run degrades to search-card-level delivery (the SUMMARY's degradedItems tells you exactly how many rows that was), rows without detail data are passed through rather than silently dropped — you get the maximum deliverable data and full visibility into how much of it was filterable.

---

Output

Each product in the dataset contains enriched fields for product research:

{
  "offerId": "951309417091",
  "titleCn": "工厂直销宠物运动腰包跑步多功能狗狗牵引绳防爆冲可伸缩免提狗绳",
  "titleEn": "Factory Direct Pet Sports Belt Running Multi-functional Dog Leash Anti-pull Retractable Hands-free Dog Rope",
  "detailUrl": "https://detail.1688.com/offer/951309417091.html",
  "images": [
    "https://cbu01.alicdn.com/img/ibank/O1CN01ObqAAw1wGklHQWemv_!!2215585976281-0-cib.jpg_270x270xzq60.jpg"
  ],
  "videoUrl": "https://cloud.video.taobao.com/play/u/2215585976281/p/2/e/6/t/1/526449776190.mp4",
  "price": {
    "min": 32.5,
    "max": 38.5,
    "currency": "CNY",
    "unit": "piece"
  },
  "quantityPrices": [
    { "price": 38.5, "minQuantity": 1 },
    { "price": 34.5, "minQuantity": 200 },
    { "price": 32.5, "minQuantity": 500 }
  ],
  "moq": 1,
  "orderCount30d": null,
  "salesTotal": 100,
  "transactionCountText": "100+ transactions",
  "repurchaseRate": 0.32,
  "categoryPath": "宠物/宠物用品/狗用品",
  "categoryEn": "Pet Supplies (All)",
  "tags": [],
  "stock": {
    "total": 20891,
    "skus": [
      { "skuId": "5928199428364", "spec": "Royal Blue > 2.5CM*(145-200)CM", "count": 9987, "price": null }
    ]
  },
  "skuDetails": [
    {
      "skuId": "5928199428364",
      "spec": "Royal Blue > 2.5CM*(145-200)CM",
      "price": null,
      "discountPrice": null,
      "stock": 9987,
      "weight": null,
      "dimensions": { "length": null, "width": null, "height": null }
    }
  ],
  "descriptionImages": [
    "https://cbu01.alicdn.com/img/ibank/O1CN01eB3IQa1wGklHR3kG7_!!2215585976281-0-cib.jpg"
  ],
  "productFlags": {
    "labels": [],
    "tags": [],
    "guarantees": [
      {
        "serviceName": "7-Day Return",
        "description": "Seller supports 7-day no-reason returns. Custom orders excluded.",
        "serviceCode": "qtwlybt"
      }
    ],
    "supplierType": "factory",
    "specialService": null,
    "isOnePiece": true
  },
  "shipping": {
    "postFee": null,
    "deliveryTime": "Ships within 48h",
    "deliveryLimitText": "48h shipping commitment",
    "sendAddress": "广东省深圳市"
  },
  "dropship": {
    "enabled": true,
    "orders30d": 156,
    "pickupRate48h": 0.86,
    "distributorCount": 23
  },
  "supplier": {
    "companyName": "深圳市一宠科技有限公司",
    "shopUrl": "",
    "cardType": "factory",
    "scores": {
      "service": "4.5",
      "repeatRate3m": "66.59%",
      "composite": 100
    },
    "saleNum": "3700+",
    "saleCountDate": "within 1 year",
    "rateInfo": {
      "positiveRate": 100,
      "score": 5
    },
    "province": "广东",
    "sellerType": "superFactory",
    "yearsOnPlatform": 4,
    "isSuperFactory": true,
    "isFactoryInspected": false,
    "city": "深圳市"
  },
  "dropshipReadyScore": 82,
  "dropshipGrade": "B",
  "scoreConfidence": "ok",
  "dataQuality": "partial",
  "translationMethod": "dictionary",
  "estLandedCostUsd": {
    "unitCostUSD": 4.51,
    "internationalShippingUSD": 2.99,
    "estimatedDutyUSD": 0.14,
    "platformFeeUSD": 0.92,
    "platformFeePercent": 12,
    "dutyPercent": 3,
    "totalLandedCostUSD": 8.56,
    "suggestedRetailPriceUSD": 21.4,
    "estimatedMarginPercent": 60,
    "markup": 2.5,
    "fxRate": 7.2,
    "currency": "USD"
  },
  "compliance": {
    "materialCategory": "general",
    "targetMarket": "US",
    "suggestedCertifications": ["CPSC compliance"],
    "riskFlags": [],
    "supplierCertificates": [],
    "disclaimer": "These hints are heuristic and for informational purposes only. Always consult a compliance expert and verify current regulations before listing."
  },
  "sourceKeyword": "dog leash",
  "scrapedAt": "2026-08-13T06:12:44.644Z"
}

Note: titleCn, supplier.companyName, shipping.sendAddress, and categoryPath retain their original Chinese text as they are raw source data from the wholesale platform. All enrichment fields (scores, costs, compliance, translations) are in English.

Field-semantics notes (0.3.39):

• dropshipStatus: null | "unknown". null (field absent or null) = no caveat — the row was either verified dropship-enabled or the filter was off; "unknown" = the item was delivered without a verifiable dropship verdict (fail-open under dropshipOnly: true).
• stock.total / stock.skus[].count: null means unknown — the platform sometimes reports sentinel values above 10,000,000 that are not real inventory counts; those are replaced with null rather than a fabricated cap.
• city vs supplierCity / supplier.city: both are kept on purpose — top-level city is the full administrative location string (province + city, kept in Chinese) straight from the detail page, while supplierCity / supplier.city is the supplier-card city translated to English where known. Use supplier.city for filtering display and city for the raw source value.
• scoreConfidence: "ok" = the dropship score rests on enough real data; "insufficient_data" = most scoring factors were missing (the grade was downgraded one level accordingly).

What a search-card-level (dataQuality: "minimal") row looks like: when an item is delivered at search-card level (see the SUMMARY diagnostic section below), it never reached the protected detail layer, so it carries only what the search results expose: title(s), image(s), headline price range and supplier card. Its quantityPrices is [], moq is null, shipping is null, and dropshipStatus is "unknown" (plus reduced enrichment fields). These are intentionally degraded research rows, not broken records — the SUMMARY's qualityMix counts exactly how many of each quality tier you received.

SUMMARY diagnostic fields

Every run ends with a SUMMARY record (key-value store) that reports exactly how the detail layer performed. Product-detail fetching is the platform's most heavily protected layer, so the Actor budgets it deliberately and degrades gracefully instead of failing silently. The key fields tell you what happened:

Field	Meaning	What to do
detailStatus	Overall detail-layer outcome for the delivered dataset: full (every delivered item has fetched details), partial (some items have details, others were delivered at search-card level), card_only (items delivered, but none got details), or a cause code (no_token, circuit_open, skipped_by_input).	full needs nothing. On partial/card_only, see the maxResults guidance below.
detailQuotaExhausted	Present (true) when the run hit its per-run detail-call budget (the platform starts blocking detail requests after a cumulative call cliff, so the Actor stops before the cliff instead of burning every remaining request). In legacy mode (fullDetailSharding: false) the leftover products are still delivered, but at search-card level. With full-detail sharding (default for large orders — see below) the run stops honestly instead of padding with cards.	Large orders shard automatically (see "Full-detail sharding" below); use fullDetailSharding: false only if you explicitly prefer the legacy card-level fallback.
degradedItems	Present only when at least one delivered item did NOT get a fetched detail layer (count of search-card-level deliveries). Its absence certifies an all-detail run.	Compare against your maxResults; combined with detailQuotaExhausted it tells you exactly how many items the budget covered.
qualityMix	Breakdown of delivered items by per-item data quality: full (detail layer + all enrichment), partial (details fetched but some fields missing), minimal (search-card-level delivery only).	Use it to judge dataset depth at a glance — e.g. { full: 100, partial: 0, minimal: 0 } is a perfect all-detail run.
detailQuota	Usage of the detail-call budget: calls (detail calls consumed) vs quota (the budget cap — in the default legacy mode 55 for orders ≤10 items and 85 for 11–55 item orders since 0.3.38; a 240 cost cap in detailIdentityRotation mode; min(maxResults × 5, 550) under full-detail sharding).	Informational — calls reaching quota is what triggers detailQuotaExhausted.
cliffDetected	Present (true) when the platform started actively blocking detail requests in this run (consecutive fully-blocked groups), so the Actor stopped launching details early to avoid burning requests.	Rerun later or rotate proxies; combine with skipDetails if you only need search-card rows.
partial + partialReason	Present (true) when the run ended early for an operational reason, regardless of detail coverage. Reason codes: deadline_expired (the run's time budget ran out before all pages were fetched), detail_quota_exhausted (the per-run detail-call budget was spent), search_supply_exhausted (the search layer could not supply enough candidates). Sharded full-detail runs use specific reasons: sfd_deadline_floor, sfd_hard_call_cap, sfd_zero_delivery, insufficient_pool, identity_pollution_wall, identity_mint_fuse (0.3.38: shard-level mint pollution persisted after a proactive remint) (0.3.40: only emitted after the bounded cooldown retry is spent). A fully filled run carries NO partialReason — shardStats.stopReason: "filled" is the success outcome.	partialReason names the cause; the flushed dataset is still complete for everything fetched up to that point. Transient platform-side blocking of the Actor's browsing sessions — not a data problem. Rerun the Actor: a fresh run starts with clean sessions. You are billed only for the rows delivered in each run.
shardStats	Present ONLY on full-detail-sharding runs: shards (shards executed, including shard-0), callsPerShard (detail calls per shard), mintPollutions / mintFailures (identity supply health), retainedParks (0.3.38: how many times the continuation-search page was reserved mid-run instead of closed — a count > 0 certifies the page-turning fix engaged), cardsDiscarded (parked search cards discarded to honour the full-detail contract), seedFallbacks (always 0 — SFD never falls back to seed tokens), stopReason (filled or the exit code above).	Informational — stopReason: "filled" certifies the quota was filled entirely with full-detail deliveries.

Rule of thumb: orders up to ~55 items fit one identity's detail budget — since 0.3.38 that budget is 55 detail calls for ≤10-item orders and 85 for 11–55-item orders (the full-detail sharding auto-ON threshold stays at 55 items — a separate switch from the per-identity call budget); larger orders automatically enable full-detail sharding (below) and keep every delivery full-detail. Under aggressive filtering (e.g. dropshipOnly culling most of a keyword's products) the delivered total can still fall short of maxResults. The SUMMARY fields above (shardStats, detailQuotaExhausted, qualityMix, partial) say exactly what happened — never guess, check them.

Full-detail sharding (large orders)

The contract: whatever you order — 1, 10 or 100 items — every delivered row carries a fetched detail layer. Never a minimal-card filler.

One fresh browser identity gets a detail budget of ~55 calls, so orders above ~55 items cannot be served by a single identity. When maxResults > 55 (or fullDetailSharding: true), the run shards automatically: shard 0 rides the warm token chain, and every further shard mints a FRESH identity with its OWN ~55-call quota, drawing survival-adjusted batches from the card pool until your quota is filled. The run's time budget is raised accordingly (up to 780s of work inside the 900s platform timeout).

Timeout note for large orders: the Actor's platform run timeout is set to 900 seconds (15 minutes) — the 780s work budget plus a 48s finish reserve need it. If you launch runs through your own API/schedule settings, keep the run timeout at 900s or higher; a lower timeout kills large orders mid-shard (the flushed partial data is still delivered and billed per row, but you will need a rerun).

Order	What to expect
≤ 55 items	Single-identity run, unchanged (~90s).
100 items	Typically 3–4 fresh identities with dropship-only filtering (2–3 without); typically ~7–9 minutes, platform cost ≈ $0.12–0.16 (compute + proxies).
Very large orders	A hard cost hat of min(maxResults × 5, 550) detail calls caps the spend. Orders above ~250 cannot be fully filled within one run's 780s budget — expect capped delivery (~110–250 rows at maxResults=500 depending on survival rate); the SUMMARY states the reason.

Honest shortfall: if the shard loop cannot fill the quota — the time floor runs out, the search pool dries up, or the platform starts poisoning fresh identities — the run stops early, delivers what it has (every row still full-detail), and SUMMARY states exactly why (partial + partialReason + shardStats.stopReason). In the worst case (observed post-detail survival below ~20%) the delivered COUNT may fall short of maxResults — but no row ever loses its detail layer to pad the number. Set fullDetailSharding: false to opt out and restore the legacy behaviour (surplus delivered at search-card level).

---

💵 Pricing

Pay only for delivered rows: $0.0045 per product + a $0.01 run fee charged once at run completion and only if at least one row was delivered. Runs that deliver nothing are not charged. No hidden usage fees.

Event	Price	When it's charged
Dataset item	$0.0045	Per product row actually delivered to the dataset
Run base fee	$0.01	Once per run, at completion — only when ≥1 row was delivered

There is a $0.04 minimum charge per run that delivers at least one product. No startup fee, no monthly subscription. A run that returns nothing costs nothing — the floor and the run fee only apply when at least one product is delivered. Each run's actual cost is reported in its SUMMARY (estimatedCost).

Example run costs:

Products	Cost
0 results (nothing delivered)	$0.00
1 (hits the $0.04 floor)	$0.04
10 (≈$0.045 rows + $0.01 run fee)	$0.06
50 (test run, ≈50 full details)	$0.24
100 (full-detail sharding — every row detailed, ~7–9 min)	$0.46
500 (niche scan; full-detail sharding — every delivered row detailed; delivery is capped by the 780s budget, see "Very large orders" above)	$2.26

Cost formula per run: products = 0 → $0.00; otherwise max($0.04, products × $0.0045 + $0.01 run fee) — the floor applies only when at least one product is delivered. The examples above use exactly this arithmetic.

Run outcomes

Every run ends in one of three states, all verifiable in the run's key-value store (SUMMARY and OUTPUT records):

Outcome	What happens	Cost
Products delivered	Dataset rows + SUMMARY report as usual	$0.0045 per delivered product + $0.01 run fee, floored at $0.04
Zero results	The run finishes SUCCEEDED with an empty dataset; SUMMARY reports the searched keywords and diagnostics	$0.00
Empty input	Both keywords and offerIds empty → the run finishes SUCCEEDED with errorCode: "NO_INPUT" in the SUMMARY and OUTPUT records, nothing researched	$0.00

Note the large-order example: with full-detail sharding (default above ~55 items) a 100-item run keeps EVERY row full-detail at the price of a longer run (~7–9 min, ≈$0.12–0.16 of platform cost). If you only need search-card rows, set skipDetails: true to skip the detail layer entirely and run faster; set fullDetailSharding: false to restore the legacy card-level fallback for the surplus.

Free tier: $5 monthly platform credit covers roughly 1,000 products ($5 ÷ $0.0045 ≈ 1,111 rows; slightly fewer across multiple runs due to the per-run fee and the $0.04 floor).

---

FAQ

Q: How much does it cost to scrape 1688? A: Pay only for delivered rows: $0.0045 per product, plus a $0.01 run fee charged once at run completion and only if at least one row was delivered, floored at a $0.04 minimum charge when at least one product is delivered, with no startup fee (see Pricing above). Runs that deliver nothing are not charged. There is no monthly subscription for the Actor itself, and the $5 free Apify credit covers roughly 1,000 products to start.

Q: Do I need a 1688 account or a Chinese phone number? A: No. The Actor accesses publicly available product data on 1688.com through Apify residential proxies. No 1688 account, no Chinese phone number, no Chinese-language skills required.

Q: Does this 1688 scraper work outside China? A: Yes. For production runs, the Actor uses Apify RESIDENTIAL proxies located in China to access 1688. During local development, it uses your direct IP.

Q: How accurate is the Dropship-Ready Score? A: The score is a weighted heuristic based on 7 factors (dropship support, MOQ, 48h pickup rate, distributor count, supplier score, 30-day orders, repurchase rate). It is a starting point — always verify before committing.

Q: How accurate is the landed cost estimate? A: It uses a fixed FX rate (1 USD = 7.2 CNY, MVP) and estimated shipping/duty/platform-fee coefficients. For production, integrate a live FX API and real shipping quotes.

Q: Can I get pet food data? A: Pet food is excluded by default (excludeFoodProducts: true) because it requires complex import permits and labeling. Set it to false if you understand the compliance requirements.

Q: 1688 changed its site — will the scraper break? A: The Actor is actively maintained and updated whenever 1688's internals change. You don't need to do anything on your side — just keep using the latest version.

---

Compliance & Disclaimer

This Actor is a data research tool that retrieves publicly available product information from China's wholesale platforms. It does not provide legal advice.

• Compliance hints are heuristic estimates based on product titles and categories. Always consult a qualified compliance expert before listing products.
• Landed cost estimates are approximations using fixed coefficients — verify with your freight forwarder and marketplace.
• Users are responsible for complying with all applicable laws, including product safety regulations (FDA, FCC, CE, Prop 65, etc.), import/customs regulations, and platform terms of service.
• The Actor operators are not liable for decisions made based on this data.

---

Tech Stack

• Runtime: Node.js 20 + ES Modules
• Framework: Apify SDK v3 + Crawlee
• Container: apify/actor-node:20-beta

Related Actors

• 1688 Image Search Scraper — reverse image search on 1688: upload a product photo, get visually similar wholesale listings with factory prices, MOQ and supplier trust signals, plus optional full-detail enrichment (SKUs, landed cost, dropship score).

Recommended pipeline: discover by keyword with this Actor (market research, category dives, winner shortlists) → source by photo with the Image Search Scraper (find the factory behind each winning product) → re-check finalists via its offerIds detail-only mode.

Reviews

Found a bug or have a wishlist? Open an Issue — a fixed bug helps you more than a one-star rating, and we answer fast.

If this run saved you time, a quick review on the Store page helps other buyers find it — 10 seconds, hugely appreciated.

License

MIT