Discuss a project

DL-036 · Case study

Price-history scoring of book ASINs for a bookseller

A bookseller needed to know which titles would sell at a profit before buying stock. We scored book ASINs from their price history and moved its in-house systems onto SP-API.

Client
Name withheld
Location
US South
Industry
Online bookseller
Capabilities
03Build02Stabilize
01

Challenge

  • Buying decisions for book stock rested on manual checks of each title.
  • Price history existed, but nothing turned it into a buy or skip signal.
  • Several internal systems still talked to Amazon through MWS.
02

Environment

  • Very large book catalogue measured in ASINs, not titles on a shelf.
  • Third-party price-history data for each ASIN.
  • In-house bookstore systems submitting listings and prices to Amazon.
03

Intervention

  • Built a scoring script that predicts profit per ASIN from price history.
  • Ran scoring in batches across the catalogue rather than title by title.
  • Added an inventory repricer driven by current market prices.
  • Moved the in-house systems from MWS calls to SP-API feeds.
04

Outcome

  • Each ASIN gets a profit score the buying team can sort and filter.
  • Listing and price updates flow through SP-API after MWS was retired.
  • Pricing follows the market instead of manual edits.
Price history per ASIN feeds a scoring engine that ranks books by predicted profit for the buying team; a repricer and SP-API feeds keep listings priced on Amazon.scoreswas MWSPRICE HISTORYPer ASINBOOK CATALOGUEASIN listSCORING ENGINEPredicted profitBUYING TEAMRanked ASINsREPRICERMarket pricesIN-HOUSE SYSTEMSListingsSP-API FEEDSPrices / stock
  1. 01PRICE HISTORYPer ASINSCORING ENGINEPredicted profit
  2. 02BOOK CATALOGUEASIN listSCORING ENGINEPredicted profit
  3. 03SCORING ENGINEPredicted profitBUYING TEAMRanked ASINsscores
  4. 04REPRICERMarket pricesSP-API FEEDSPrices / stock
  5. 05IN-HOUSE SYSTEMSListingsSP-API FEEDSPrices / stockwas MWS
Fig. 01From price history to a buy decisionPrice history per ASIN feeds a scoring engine that ranks books by predicted profit for the buying team; a repricer and SP-API feeds keep listings priced on Amazon.Scale NTS

Stack and services.

Stack06

  • Amazon SP-API
  • SP-API Feeds
  • Price-history API
  • Batch scoring jobs
  • Repricing engine
  • Amazon MWS (retired)

Services used02

Tell us what has to work.

A senior engineer reads every brief. Describe the system, what it runs on today and what happens when it fails.

Discuss a project