Newegg.com - Electronics Retail Product Catalog (Scrapeable)

The other anchor of American online PC retail, opened up as structured records instead of page furniture. Newegg sells GPUs, CPUs, motherboards, memory and storage, full systems, peripherals, servers and AI hardware, networking gear, gaming equipment, appliances and smart home products across roughly sixteen top-level departments - and unlike most retailers its listing pages carry a machine-readable product object behind every card rather than prices that exist only as rendered pixels.

Scale and depth together: hundreds of thousands of live listings sit under those departments, resolved into hundreds of leaf subcategories ('GPUs / Video Graphics Cards', 'Desktop CPU Processor', 'Gaming Desktop PC'), and each card's embedded payload is unusually complete for a retailer - not just a price but a was-price, an instant-rebate amount and a trailing-30-day price low on the same row. During the August 2026 research pass, one deals page alone yielded 21 fully-populated item records, about 196 KB of product JSON.

This record sits in the realtime tier of Datadory's catalog - only 197 of the 1,744 datasets we catalog update in real time - which matters here because promotions turn over constantly. Datadory normalizes those payloads into the typed rows below, delivered daily, weekly, or hourly. [Get a sample of this dataset](#request) and judge the columns.

One row per listing, exactly as it lands in your warehouse. Values below are verified example values carried on the field definitions:

item                : 14-932-827
title               : GIGABYTE Gaming Radeon RX 9070 GRE ... OC-12GD
finalPrice          : 534.99         unitCost : 567.49
instantRebate       : 32.50          lowestPrice30Days : 529.64
instock             : true           rating : 4.5   reviews : 235
brand               : GIGABYTE       model : GV-R907GREGAMING OC-12GD
subcategory         : GPUs / Video Graphics Cards

item                : 19-118-628     title : Intel Core Ultra 7 270K Plus ...
finalPrice          : 310.00         unitCost : 330.19
subcategory         : Desktop CPU Processor   brand : Intel

item                : 9SIC7XVM3K2601 (marketplace key)
title               : Mloong Gaming PC Desktop, Ryzen 7 8700F / RTX 5060 Ti ...
finalPrice          : 1279.00        unitCost : 1799.00
seller              : MLOONG DIGITAL  subcategory : Gaming Desktop PC

Read it as three different jobs in one schema. The first row is a repricing kit: final against unit-cost reference plus the rebate amount makes discount depth arithmetic, and the 30-day low tells you whether today's sticker is genuinely cheap. The second is a component panel keyed by brand and model for vendor-side joins. The third shows the marketplace dimension - a longer alphanumeric item key and a named seller - which first-party rows simply leave empty.

Twelve documented field groups anchor the dictionary, all verified against live listing payloads during research - these definitions come off real cards, not documentation guesses. Deeper attributes observed on the same objects (line and bullet description variants, shipping-day estimates, launch dates, weights and dimensions, quantity limits, condition flags) are held under additional fields on request rather than promised in every row.

The pricing quartet is the distinctive part. Most retailer feeds give a price and maybe a list price; this one separates the pre-discount reference (UnitCost), the dollars actually knocked off (InstantRebateAmount) and the regulatory-style floor (LowestPrice30Days), so promotional mechanics can be reconstructed rather than inferred.

Geography - the US storefront catalog. Regional Newegg sites exist for Europe, South America, Asia Pacific and the Middle East, but this record covers the US assortment only. If your benchmark needs a single consistent currency and taxonomy, that constraint is a feature.

Temporal - a living catalog. Prices, stock and promotions change continuously and the source publishes no archive, so any longitudinal view comes from scheduled captures on a cadence you choose. The LowestPrice30Days field partially compensates: every snapshot carries a rolling floor price, so even a single capture knows something about the prior month.

Granularity - one record per listing, organized by roughly 16 top-level departments down to hundreds of leaf subcategories. Deal and event pages expose curated subsets of the same schema, which makes them useful high-signal slices: what appears there is what the merchant is actively pushing.

API, files, or your warehouse. Daily, weekly, or hourly.

Hourly suits Shell Shocker-style promotion windows, where a deal price can appear and vanish inside a day. Daily suits competitor tracking on PC-component categories, where list prices move weekly but rebates churn faster. Weekly suits quarterly benchmarking and assortment studies where the trend line outranks any single sticker. Whichever cadence you pick, the join keys hold still: the Newegg item number keys the row internally, brand plus model strings tie it back to vendor line items and to overlapping SKUs at other retailers.

• PC-component price and rebate panels - the FinalPrice/UnitCost/InstantRebate trio turns GPU, CPU and memory promotions into comparable series across brands; mapped to the price monitoring use case.
• Marketplace seller intelligence - Seller.SellerId, SellerRating and SellerReviewCount quantify who competes with the merchant on its own shelves, and at what discount; see the competitor tracking use case.
• Condition-aware assortment analysis - IsRefurbished, IsOpenBoxed and IsNew flags let analysts separate genuine new-unit pricing from graded-stock discounts instead of averaging them together.
• Demand signals for hardware models - review averages and counts beside price moves give SKU-level demand features for consumer-electronics forecasting; pairs naturally with federal retail aggregates in the demand forecasting use case.
• Catalog mapping and enrichment - item numbers, brands, models and category paths normalize a messy reseller feed onto a clean reference taxonomy.
• Gaming-hardware market studies - the subcategory tree isolates GPUs, CPUs and gaming desktops precisely, the categories journalists and analysts quote most.

E-commerce operators get the repricing reference for PC hardware, where shoppers treat Newegg prices as the market rate for components. Competitive intelligence and product teams track rebate depth, stock state and seller mix on a rival whose catalog is the enthusiast segment's price index. Investors and quants read component pricing and review velocity as a demand signal for semiconductor and gaming cycles between earnings prints. Data scientists and ML engineers get typed price, brand, rating and condition panels clean enough for elasticity work without scraping cleanup. Developers building data products get stable item-number keys and documented examples, the difference between an integration and a science project. Market researchers and consultants benchmark PC-hardware assortment breadth by subcategory without assembling it by hand.

Three things worth knowing upfront. First, this is a live-listing record: the source keeps no archive, so a historical backfill is impossible - anything longitudinal must be built forward from scheduled captures. Second, the card payload carries no UPC field, so cross-retailer joins run through brand-plus-model string matching rather than a universal code; budget normalization time accordingly, or pair this feed with the Best Buy catalog where a UPC rides on every SKU. Third, marketplace listings are a meaningful share of rows and behave differently - longer alphanumeric item keys, seller-rated offers, wider price dispersion - so decide before sampling whether you want first-party rows only or the full shelf.