Datasets
Buyer room
PRICEFIND.APPManage supplier listing
PriceFind (pricefind.app) — Crowdsourced in-store grocery shelf-price observations, each anchored to a shopper-submitted photo of a shelf-edge price tag. Per observation the app digitally captures the barcode (UPC or store SKU), GPS-stamps the specific store, and uses a vision model to read product name, price, quantity, unit and any sale/promotion detail off the tag. Every record is time-stamped and carries image-quality scores (sharpness, skew, a live 'focus score' at capture) and validation results (UPC checksum, price-per-unit arithmetic). Supporting layers: a store registry with entrance…
supplier index read from their site
Crowdsourced in-store grocery shelf-price observations, each anchored to a shopper-submitted photo of a shelf-edge price tag. Per observation the app digitally captures the barcode (UPC or store SKU), GPS-stamps the specific store, and uses a vision model to read product name, price, quantity, unit and any sale/promotion detail off the tag. Every record is time-stamped and carries image-quality scores (sharpness, skew, a live 'focus score' at capture) and validation results (UPC checksum, price-per-unit arithmetic). Supporting layers: a store registry with entrance photographs and coordinates, contributor earning rates and trust tiers, a per-item star rating, an error-report queue, recipe-to-item ingredient matching, and publication freshness gates (shown as a current price plus Offer structured data within 14 days, retained with an 'observed N days ago' note at 15-30 days, suppressed beyond 30 days). Explicitly NOT retailer-site scraping — the data is first-party physical-shelf capture.
Coverage Consumer Staples · Information TechnologyAsset class Equities · FxTickers WMT · COST · KR
Sample, licence terms, pricing and eval results when Pricefind publishes them. Until then, discover alternatives today with a 7-day trial.
Start 7-day trial →What buyers ask about Pricefind, answered from this page.
PriceFind (pricefind.app) — Crowdsourced in-store grocery shelf-price observations, each anchored to a shopper-submitted photo of a shelf-edge price tag. Per observation the app digitally captures the barcode (UPC or store SKU), GPS-stamps the specific store, and uses a vision model to read product name, price, quantity, unit and any sale/promotion detail off the tag. Every record is time-stamped and carries image-quality scores (sharpness, skew, a live 'focus score' at capture) and validation results (UPC checksum, price-per-unit arithmetic). Supporting layers: a store registry with entrance…
Pricefind offers (Alternative, Price, Sentiment, Fundamental) — Not disclosed and the dashboard metrics render blank for logged-out visitors. Bounded by the mechanic rather than the counter: each observation carries one or more photographs plus a structured extraction, so volume should be quoted as images-plus-records, not rows. Order of magnitude: a zone-limited contributor programme of hundreds of shoppers submitting per-shop captures would plausibly accumulate tens of thousands to a few hundred thousand tagged observations, with a high price of re-observation for popular UPCs (repeat observations of the same item are the norm, since the site gates each product page on its most recent observation). Ask for the exact observation and image counts, and for the unique-UPC-by-store grid separately from raw observation count — the two scale very differently..
Primary buyer set is not retail analytics — it is AI training data. (1) OCR / document-understanding and receipt-or-shelf-tag extraction model training and evaluation, with verified ground truth and human-corrected error cases. (2) Vision-model grounding and structured-extraction benchmarks for price tags, a documented failure mode of general multimodal models. (3) Image-quality and focus/sharpness scoring model development, using the built-in readability scores. (4) Retail media and CPG analytics: true shelf price and promotion tracking per UPC per store, independent of retailer-published web prices — the offline channel is exactly what retailer-site scraping misses. (5) Inflation and price-dispersion research, including store-to-store variance for the same UPC at the same date. (6) Grocery deal-finding and recipe-basket pricing products. (7) Storefront image classification from the entrance-photo library.
Coverage spans US, UK, Other; Food Retail, Data Processing & Outsourced Services, Drug Retail; alternative, price, sentiment, fundamental; equities, fx.
Pricefind has not added their own details yet. Not yet on file: