Datadory notebook
Electronic components price history database: what exists, and how a curve gets built
Datadory delivers electronic components price history covering both halves of a working price database: part-level price-and-stock frames built from repeated capture across four named distributor shelves - Nexar API (Octopart), FindChips Aggregated Price & Inventory Search, the LCSC Electronics Catalog and TME's European warehouse - keyed on one manufacturer part number, joined to the market anchors that explain the moves: UN Comtrade Plus implied unit values per HS 85/90 line, Federal Reserve G.17 production indexes for NAICS 334 reaching 1972, and the UK ONS Index of Production's Division 26 series. Delivered daily, weekly, or hourly - your call.
1,744 datasets. Pick your catch.
Where does the history come from if no feed ships it?
Rank the watchlist before setting the schedule. Three signals sort it: sole-source exposure, where one maker or one channel stands between a design and a stop-ship; lifecycle divergence, where the flag says active and the offer count says otherwise; and BOM weight, where a penny part on ten million units outranks a module bought twice a year. The 791-row supply-chain risk screen adds origin concentration - China-origin flags with confidence scores and reason codes - so geographic exposure enters the ranking before design freeze rather than after.
Why do four shelves beat one?
Because the shelves disagree, and the disagreement is the analysis.
Can customs data stand in for part-level history?
At product-line resolution, yes - and it is the only long-run proxy in the pool. UN Comtrade Plus TradeFlow - Electrical Machinery & Instruments (HS 85/90) reports bilateral flows as reporter x partner x six-digit HS line x period, each carrying dollar value, weights, supplementary quantities and explicit estimation flags. Divide value by weight or quantity and an implied average unit value falls out per line per partner: the nearest thing this industry has to a published component price index, reaching back to 1962.
How do captures become a database that survives modeling?
Accumulation is easy; surviving contact with a model is the work. Five properties decide it.
- The observation date outranks the retrieval date. Every row carries the day it was true. Joins use that column - an August capture of a May board describes May, whatever arrived when.
- Units stay beside amounts. Dollars and yuan coexist in one frame, as do per-unit and per-ladder figures. A join key without the currency column silently mixes markets that were never priced alike.
- Both grammars get normalized. The Nexar shape nests offers beneath one MPN; FindChips emits one row per offer; TME runs one row per product symbol. Collapse all three onto MPN x channel x capture date and the stack becomes one queryable table.
- Join keys are declared, not discovered. MPN joins the components layer to CAD libraries and the risk screen; CDID joins ONS series; reporter-partner-commodity-period joins trade flows. Cross-dataset work needs no fuzzy matching pass.
- Provenance rides on every record. Aggregate flags, estimation flags and lifecycle verdicts travel with their rows, so a downstream model can exclude what it cannot defend.
Where to go next
This page covered the part-level price history frame and the market anchors that explain it. Continue with:
- Nexar API (Octopart) - the dataset product page behind the part-level layer, with the full field dictionary, sample rows and coverage chips.
- Component lifecycle status EOL data - the obsolescence watch that tells you which parts deserve a price history in the first place.
- Live Component Price & Stock Data: The Four Feeds - the live layer compared record by record.
- Japan Consumer Electronics Shipment Statistics - the demand-side read that leads commodity component orders.
- Electronic Equipment & Instruments Data Guide - the pillar mapping all 24 pooled records in the instruments slice.
For the structured index behind all of it, the electronic components data hub lists every cataloged record with coverage, granularity and use cases - start a price history before you need the curve.
| Field | Type | Definition | Example |
|---|---|---|---|
| mpn | string | Manufacturer part number identifying the component; the join key across every layer of the price history frame. | ESP8266EX |
| manufacturer | string | Component maker paired with the MPN; together the two form one canonical record identity. | Espressif Systems |
| capture_date | date | The date a snapshot was taken - the column that turns repeated deliveries into a series. | 2026-08-14 |
| channel | string | The named selling shelf quoting the part within the aggregated distributor network. | DigiKey |
| unit_price | number | Price per unit at a stated quantity break, with the currency declared beside it. | 1.45 USD at quantity 1,000 |
| price_breaks | array | Nested quantity-step ladder per offer: price, currency and quantity together. | [{"quantity": 10, "unit_price": 1.38, "currency": "USD"}] |
| stock_level | integer | Quantity available at the quoted channel on the capture date. | 6433 |
| lifecycle_status | enum | Manufacturing lifecycle verdict carried as a filterable field rather than datasheet prose. | Active |
| factory_lead_days | integer | Estimated factory lead time in days riding on the part record. | 42 |
| implied_unit_value | number | Derived column for trade rows: dollar value divided by reported weight or quantity, per HS line and partner. | 1723 USD/kg |
Pick up where this leaves off
Every one of these ships with sample rows before you commit to anything.
Nexar API (Octopart) Data
Octopart Electronic Components Search & BOM Tool
mpn · manufacturer · description …+5 more
FindChips Aggregated Price & Inventory Search
LCSC Electronics Catalog
TME Electronic Components Catalog & API
symbol · mpn · country …+6 more
Electronic Components Supply Chain & Risk Dataset
Want rows instead of a pitch? Name the datasets.
API, files, or your warehouse. Daily, weekly, or hourly.
Get a sampleQuestions worth asking
How far back does electronic component price history go?
Two answers, two resolutions. Part level: as far back as your capture window plus whatever trend views the operator retains - weekly capture yields a usable curve within a quarter. Market level: decades. UN Comtrade Plus bilateral flows reach back to 1962, Federal Reserve G.17 production indexes for NAICS 334 to 1972, and the Observatory of Economic Complexity cube spans 1995-2024.
Which parts belong in a price-history watchlist first?
Three signals rank it: sole-source exposure, lifecycle divergence - a part flagged active while its offer count thins is the earliest discontinuation symptom in the stack - and BOM weight, where cheap parts on huge volumes outrank expensive ones bought rarely. Origin concentration from the 791-row supply-chain risk screen belongs in the ranking before design freeze.
How is component price history delivered?
As files, scheduled feeds, or a direct warehouse write - daily, weekly, or hourly, your call. Rows arrive flattened to one observation per part per channel per capture date, with currencies, units and provenance flags beside their values and both nested and flat grammars pre-normalized. A sample scoped to your MPN list ships first, dictionary attached.