Methodology
How we collect this data
Sneakerblend combines marketplace data, retail observations, search interest, social observations and its own research. This page names every source, how often it is collected and what each number measures.
- 01
Collect
Daily sales and asking prices for the tracked list, retailer observations, Google Trends exports and hand-logged posts.
- 02
Link
Every record is tied to a brand, model, colour and material.
- 03
Review
An editor checks the source and the links.
- 04
Publish
Members see published records only. History is never overwritten.
On this page
Where it comes from
Sources and cadence
- Marketplace asking pricesDaily, 03:00 UTC
- Current asking prices per size from StockX, received through the KicksDB API. KicksDB refreshes its data at least once every 24 hours; our daily sync stores each snapshot and never overwrites an earlier one, so the price history grows day by day.
- Coverage now: 574 tracked SKUs, last synced 15 Nov 2026.
- Marketplace sales historyDaily, 03:00 UTC
- Completed sales from two marketplaces, StockX and GOAT, received through the KicksDB API. StockX gives the daily average sale price and order count per product; GOAT gives each individual sale of a tracked sneaker, in US dollars.
- Coverage now: 98 weeks of history across 574 tracked SKUs.
- Bestseller rankingsWeekly
- The StockX and GOAT bestseller rankings, sneakers only (top 1,000 on each), saved once a week through the KicksDB API. The 100 sneakers with the most combined weekly orders are followed daily.
- KREAM (Korea, beta)Weekly
- KREAM's lifetime sales total for tracked sneakers, saved once a week and shown with the date of the save. KREAM reports only a lifetime total, so the weekly change is the total on the latest save minus the total on the save at least 6 days earlier; a total that fell is left out. It is shown as a separate figure and never added to marketplace orders.
- Retail observationsAs observed
- The online assortment of a selection of retailers, recorded by hand or imported by CSV, then linked to brand, model, colour and material and reviewed before publication.
- Coverage now: 42 retailers in 12 countries, 2,357 published observations.
- Search interestWhen an export is uploaded
- Google Trends exports, downloaded from Google and uploaded by hand. Nothing is fetched automatically.
- Coverage now: 6 published reports.
- Social observationsAs observed
- Individual TikTok and Instagram posts logged by hand with their link, creator, dates and visible engagement. Nothing is scraped.
- Coverage now: 11 published observations.
- ResearchAs written
- Analysis written and reviewed in house, published with its sources and the date it was written.
- Coverage now: 4 published articles.
Asks and sales
An asking price is not a sale
An ask is what a seller wants for a pair; a sale is what a buyer paid. Every price on the platform says which one it is. Asks are shown as the lowest ask per size in the most recent snapshot. Sales are shown as an average sale price, with the number of orders behind it.
Asks and sales are never merged into one series. Where both exist they are charted separately.
Marketplace orders
Two marketplaces, one count
Marketplace orders are StockX's orders plus GOAT's sales of tracked sneakers in the same ISO week. The two are separate marketplaces, but nothing in the provider's data links a sale on one to a sale on the other, so a pair sold through both cannot be ruled out. GOAT does not say whether a pair was new or used, so orders cannot be split by condition.
Markets and currencies
Different markets are different prices
Marketplaces do not report where buyers are, so marketplace figures are worldwide: no country or European sales claim is made. StockX's sales are priced on its German basis in euros; its orders are the worldwide order stream, not German sales.
GOAT amounts are in US dollars and are converted to euros at the week's published rate. Without a rate, GOAT orders still count but stay out of the average price, and the page says so.
Asking prices come from the Dutch market, in euros. A price in one market is not a conversion of a price in another (fees, taxes and supply differ), so asks and sales are never drawn on the same axis.
Coverage
Tracked sneakers, not the whole market
Tracked sneakers are the ones SNEAKERBLEND follows every day. Broad market is the StockX and GOAT bestseller rankings, sneakers only (top 1,000 on each), saved weekly; its top 100 by combined weekly orders are followed daily. Neither is the whole sneaker market.
A provider's ranking can trail the marketplace by a few days, so each ranking shows the date its data was last updated.
Retail observations
Assortment is not sell-through
A product appearing at a monitored retailer means it was offered, not that it sold. Colour, material and category counts describe what retailers carried. Every observation keeps its retailer, date and source.
Search interest
Relative interest, normalised per export
Google Trends reports interest on a 0–100 scale, where 100 is the peak inside one export. It is not a number of searches. Because Google normalises each export separately, terms from different exports are never charted on the same axis.
Rankings
Every ranking states its method
A ranking shows the dataset it used, the period it covers, how it was calculated and the filters applied. Rankings count published observations; they describe what was observed, not what sold.
Each ranking has a minimum number of observations and distinct entries. Below it, the ranking is not drawn: the page says “Insufficient data” and shows the minimum and the current count. Emerging signals compare each attribute's share of observations with the previous period and flag a rise only when it appears across several brands and at least two kinds of source.
Images and rights
Linked, not copied
Product images are linked from the marketplace that hosts them, not copied to our servers, and remain the property of their owners. The provider's terms permit storing and displaying its data. Social posts are linked to the original post.
Review
Nothing is published unreviewed
Retail, social, search and research records arrive unpublished. An editor checks the source and the links to brand, model, retailer, colour and material before anything is published. Records are never overwritten, so the history stays intact.
About SneakerblendTerms
Glossary
- Ask
- What a seller wants for a pair at a moment in time.
- Sale (completed sale)
- What a buyer paid. Shown as the average price per day, with the number of orders behind it.
- Relative interest
- Google Trends' 0 to 100 index, where 100 is the peak inside one export.
- Assortment
- What a retailer offers at a point in time.
- Sell-through
- The share of stock that sold. We do not measure it.
- Tracked list
- The products whose prices and sales are synced daily. Chosen by an editor; not the whole market.
- Edition
- One weekly Market Pulse for one ISO week, frozen when published. A correction is a new version.
- Insufficient data
- Shown instead of a ranking when there are too few observations to rank.
Log
Method changes
- 19 Oct 2026Market Pulse weekly editions added, for the tracked list on the German market. Published editions are frozen; a correction is a new version.
- 19 Oct 2026Search reports added: each Google Trends export is published as one reviewed report.
- 18 Oct 2026Retail snapshots added: a retailer's assortment counted on a date, reviewed before publication.
- 17 Oct 2026Completed-sale history added: daily average sale price and number of orders per product, German market.
Social observations
Engagement is not popularity.
Posts are a selection, not a measurement of a platform. Brand-owned content is marked apart from creator and community posts, and likes, comments and views are shown as they were visible on the day, never as evidence of sales. A metric that was not visible is left empty.