Datasets:

Octoparse
/

temu-ecommerce-pricing-workflow-sample

	# Data Dictionary — Temu E-commerce Pricing & SKU Variant Dataset
	Version: 2.0.0 \| Provider: Octoparse Managed Data Service

	---

	## Table: `products.parquet`
	One row per Temu product listing (SPU). Contains product-level metadata.
	85 rows.

	\| Field \| Type \| Example \| Business Context \|
	\|---\|---\|---\|---\|
	\| `spu_id` \| string \| `"606068486800016"` \| Temu's Standard Product Unit ID. Use as join key with `skus.parquet`. Stored as string to prevent int64 overflow. \|
	\| `product_url` \| string \| `"https://m.temu.com/product-g-606068486800016.html"` \| Canonical mobile URL. Append to any Temu domain for desktop. \|
	\| `is_on_sale` \| boolean \| `True` \| Whether the product is currently purchasable. All 85 products in this sample are active. \|
	\| `not_on_sale_reason` \| string \| `"Unavailable for purchase"` \| Only populated when `is_on_sale=False`. Null for all active products. \|
	\| `goods_property_summary` \| string \| `"Material:Polyester;Style:Casual;Season:Summer"` \| Semicolon-delimited key:value property bag. Keys vary widely across products (~178 unique keys). Useful for taxonomy building. Parse defensively (typos present, e.g. `"Seasoon"`). \|

	---

	## Table: `skus.parquet`
	One row per SKU (product variant). Foreign key `spu_id` joins to `products.parquet`.
	382 rows, avg 4.5 SKUs per product.

	\| Field \| Type \| Example \| Business Context \|
	\|---\|---\|---\|---\|
	\| `sku_id` \| string \| `"17592356187455"` \| Temu's variant-level ID. Unique per color/size/style combination. \|
	\| `spu_id` \| string \| `"606068486800016"` \| FK → `products.spu_id`. \|
	\| `sku_main_pic` \| string \| `"https://aimg.kwcdn.com/..."` \| CDN URL for the SKU's primary image. URLs are static (not signed). Use for visual product matching or multimodal AI pipelines. \|
	\| `sku_original_price` \| float64 \| `12.99` \| Strikethrough list price in USD. Null for ~69% of SKUs — Temu only renders this when actively showing a "was/now" comparison. Absence ≠ no discount; it means Temu isn't surfacing the comparison at this moment. \|
	\| `sku_discounted_price` \| float64 \| `4.89` \| Current selling price in USD at scrape time. Range: $2.14 – $219.72. \|
	\| `discount_pct` \| float64 \| `0.6235` \| Computed: `(original − discounted) / original`. Only populated when both prices present (120/382 SKUs). Use `log1p` transform for modeling. \|
	\| `sale_attr_color` \| string \| `"Black"` \| Color variant label. May bundle quantity (e.g. `"Red Rose Vine - 3pcs"`). \|
	\| `sale_attr_size` \| string \| `"6.5-7"` \| Size variant. Format varies: numeric (`8`), range (`6.5-7`), children's age (`3-4Y`), clothing (`XL`). Normalize before comparison. \|
	\| `sale_attr_quantity` \| string \| `"1pc"` \| Quantity or bundle label. Highly variable — ranges from `"1pc"` to full product descriptions. Not reliably machine-parseable as numeric. \|
	\| `sale_attr_capacity` \| string \| `"500ml"` \| Volume/storage capacity. Sparse (<1% populated). \|
	\| `sale_attr_model` \| string \| `"Standard"` \| Product model variant. Sparse (~2% populated). \|
	\| `sale_attr_style` \| string \| `"Modern"` \| Style variant. Sparse (~1% populated). \|
	\| `sale_attr_compatible_model` \| string \| `"iPhone 15"` \| Device compatibility for accessories. Sparse (~1% populated). \|
	\| `sale_property_summary` \| string \| `"Color:Black\\|Size:XL"` \| Pipe-delimited summary of all non-null `sale_attr_*` values for this SKU. Prefer this over individual columns for display or search. \|

	---

	## Derived Metrics (compute on demand)

	\| Metric \| Formula \| Use Case \|
	\|---\|---\|---\|
	\| Discount Rate \| `discount_pct` (precomputed) \| Price competitiveness signal \|
	\| SKU Count per Product \| `COUNT(sku_id) GROUP BY spu_id` \| Variant complexity scoring \|
	\| Price Spread per Product \| `MAX(sku_discounted_price) - MIN(sku_discounted_price) GROUP BY spu_id` \| Dynamic pricing range per SKU matrix \|
	\| Has Strikethrough Price \| `sku_original_price.notna()` \| Promotion detection filter \|

	---

	## Key Data Limitations

	\| Issue \| Detail \|
	\|---\|---\|
	\| `sku_original_price` sparsity \| 68.6% null — Temu algorithmically controls when strikethrough prices are shown \|
	\| `sale_attr_*` mutual exclusivity \| Each SKU uses at most 1–2 of 7 attribute columns; remaining are null \|
	\| Price is a snapshot \| No historical cadence in this sample — prices change frequently on Temu \|
	\| `goods_property_summary` key inconsistency \| ~178 unique keys, some duplicated under different spellings \|

	---

	Generated by Octoparse Managed Data Service pipeline `octoparse-temu-pricing-v2`. For production-scale Temu price monitoring pipelines (hourly cadence, 10K+ SKUs), visit [octoparse.com/managed-data-service](https://www.octoparse.com/managed-data-service).

	# Data Dictionary — Temu E-commerce Pricing & SKU Variant Dataset
	Version: 2.0.0 \| Provider: Octoparse Managed Data Service

	---

	## Table: `products.parquet`
	One row per Temu product listing (SPU). Contains product-level metadata.
	85 rows.

	\| Field \| Type \| Example \| Business Context \|
	\|---\|---\|---\|---\|
	\| `spu_id` \| string \| `"606068486800016"` \| Temu's Standard Product Unit ID. Use as join key with `skus.parquet`. Stored as string to prevent int64 overflow. \|
	\| `product_url` \| string \| `"https://m.temu.com/product-g-606068486800016.html"` \| Canonical mobile URL. Append to any Temu domain for desktop. \|
	\| `is_on_sale` \| boolean \| `True` \| Whether the product is currently purchasable. All 85 products in this sample are active. \|
	\| `not_on_sale_reason` \| string \| `"Unavailable for purchase"` \| Only populated when `is_on_sale=False`. Null for all active products. \|
	\| `goods_property_summary` \| string \| `"Material:Polyester;Style:Casual;Season:Summer"` \| Semicolon-delimited key:value property bag. Keys vary widely across products (~178 unique keys). Useful for taxonomy building. Parse defensively (typos present, e.g. `"Seasoon"`). \|

	---

	## Table: `skus.parquet`
	One row per SKU (product variant). Foreign key `spu_id` joins to `products.parquet`.
	382 rows, avg 4.5 SKUs per product.

	\| Field \| Type \| Example \| Business Context \|
	\|---\|---\|---\|---\|
	\| `sku_id` \| string \| `"17592356187455"` \| Temu's variant-level ID. Unique per color/size/style combination. \|
	\| `spu_id` \| string \| `"606068486800016"` \| FK → `products.spu_id`. \|
	\| `sku_main_pic` \| string \| `"https://aimg.kwcdn.com/..."` \| CDN URL for the SKU's primary image. URLs are static (not signed). Use for visual product matching or multimodal AI pipelines. \|
	\| `sku_original_price` \| float64 \| `12.99` \| Strikethrough list price in USD. Null for ~69% of SKUs — Temu only renders this when actively showing a "was/now" comparison. Absence ≠ no discount; it means Temu isn't surfacing the comparison at this moment. \|
	\| `sku_discounted_price` \| float64 \| `4.89` \| Current selling price in USD at scrape time. Range: $2.14 – $219.72. \|
	\| `discount_pct` \| float64 \| `0.6235` \| Computed: `(original − discounted) / original`. Only populated when both prices present (120/382 SKUs). Use `log1p` transform for modeling. \|
	\| `sale_attr_color` \| string \| `"Black"` \| Color variant label. May bundle quantity (e.g. `"Red Rose Vine - 3pcs"`). \|
	\| `sale_attr_size` \| string \| `"6.5-7"` \| Size variant. Format varies: numeric (`8`), range (`6.5-7`), children's age (`3-4Y`), clothing (`XL`). Normalize before comparison. \|
	\| `sale_attr_quantity` \| string \| `"1pc"` \| Quantity or bundle label. Highly variable — ranges from `"1pc"` to full product descriptions. Not reliably machine-parseable as numeric. \|
	\| `sale_attr_capacity` \| string \| `"500ml"` \| Volume/storage capacity. Sparse (<1% populated). \|
	\| `sale_attr_model` \| string \| `"Standard"` \| Product model variant. Sparse (~2% populated). \|
	\| `sale_attr_style` \| string \| `"Modern"` \| Style variant. Sparse (~1% populated). \|
	\| `sale_attr_compatible_model` \| string \| `"iPhone 15"` \| Device compatibility for accessories. Sparse (~1% populated). \|
	\| `sale_property_summary` \| string \| `"Color:Black\\|Size:XL"` \| Pipe-delimited summary of all non-null `sale_attr_*` values for this SKU. Prefer this over individual columns for display or search. \|

	---

	## Derived Metrics (compute on demand)

	\| Metric \| Formula \| Use Case \|
	\|---\|---\|---\|
	\| Discount Rate \| `discount_pct` (precomputed) \| Price competitiveness signal \|
	\| SKU Count per Product \| `COUNT(sku_id) GROUP BY spu_id` \| Variant complexity scoring \|
	\| Price Spread per Product \| `MAX(sku_discounted_price) - MIN(sku_discounted_price) GROUP BY spu_id` \| Dynamic pricing range per SKU matrix \|
	\| Has Strikethrough Price \| `sku_original_price.notna()` \| Promotion detection filter \|

	---

	## Key Data Limitations

	\| Issue \| Detail \|
	\|---\|---\|
	\| `sku_original_price` sparsity \| 68.6% null — Temu algorithmically controls when strikethrough prices are shown \|
	\| `sale_attr_*` mutual exclusivity \| Each SKU uses at most 1–2 of 7 attribute columns; remaining are null \|
	\| Price is a snapshot \| No historical cadence in this sample — prices change frequently on Temu \|
	\| `goods_property_summary` key inconsistency \| ~178 unique keys, some duplicated under different spellings \|

	---

	Generated by Octoparse Managed Data Service pipeline `octoparse-temu-pricing-v2`. For production-scale Temu price monitoring pipelines (hourly cadence, 10K+ SKUs), visit [octoparse.com/managed-data-service](https://www.octoparse.com/managed-data-service).