embedded JSON product payload

Product payloads differ from generic page data in one way that matters commercially: they carry the offer. Not just name and image, but the current price, the crossed-out was-price, any rebate, and live stock state - the fields a merchandising team actually changes.

The cataloged record is precise. Newegg.com - Electronics Retail Product Catalog (Scrapeable) lists data_formats html and embedded_json, describes pages that 'embed a structured JSON product payload per item', and quantifies density: a single deals page embedded 21 fully-populated ItemCell objects totaling about 196 KB of product JSON. Coverage spans hundreds of thousands of live listings organized by ~16 top-level departments and hundreds of leaf subcategories. In Datadory's slice this source is tagged computer-electronics-retail with technology-hardware-storage-peripherals as a related industry.

Hardware price tracking fails on missing history more often than missing fields, and payload structure is where you find out.

• Was-price and rebate fields expose real discounts. Street pricing on PC components swings through rebates; a payload without was-price and rebate fields cannot distinguish a genuine cut from a coupon round-trip.
• Stock state timestamps the offer. Stock flips make the difference between quoting a price and buying at it.
• No archival price history comes with it. These fields describe the current offer; longitudinal analysis requires you to capture repeatedly - which collides with Newegg's recorded terms prohibiting automated access.
• Buyers who skip the payload check discover at integration time that a vendor's 'price data' is screenshots of the same page.

1. Enumerate the field set per object. Confirm each payload carries item number, title, price, was-price, rebate, stock, brand, model, rating and review count - the cataloged schema for Newegg's ItemCell objects.
2. Measure payload weight against density. 21 objects x ~196 KB on one deals page is your baseline; pages far below that are probably curated subsets, not the full department.
3. Map payloads onto the taxonomy. Newegg organizes ~16 top-level departments and hundreds of leaf subcategories; coverage claims should decompose along that tree, not a flat page list.

Where to see it in context: computer-electronics-retail data, technology-hardware-storage-peripherals data.

• offer-price-stock-fields - the general field family (price, was-price, rebate, stock) these payloads instantiate, which change continuously and typically carry no archival history.
• pc-components-catalog - the ~16-department e-commerce taxonomy those payloads populate, from components and systems to peripherals and appliances.