Wine-Searcher Scraper: Scores, Prices & Popularity

Look up any wine on Wine-Searcher and get a clean, structured record back. Give the actor a list of wine names, Wine-Searcher links, or LWIN codes and it returns critic scores, average and market prices, grape, region, appellation, producer, label image, user ratings and live offer counts as JSON or CSV. Built for merchants, collectors, sommeliers and market analysts who need Wine-Searcher data in a spreadsheet or pipeline instead of a browser tab.

Why this scraper

• Three input styles in one actor: wine names, Wine-Searcher URLs, or LWIN codes (LWIN7 / 11 / 16 / 18), with automatic detection so you can mix them.
• Richer output than a plain price lookup: 30+ fields per wine, including user ratings, food pairing, region and origin, that other tools drop.
• Cheapest live offer with the merchant name when you enable offer extraction, plus the total number of offers worldwide.
• Optional per-critic review breakdown (reviewer name and score) when you enable review collection.
• Looks up several wines in parallel, over several connections at once on long lists, so a few-hundred-wine list finishes in minutes rather than hours.
• Long lists keep going: the actor re-establishes its connection as often as it needs to, and a wine that could not be read on one connection is retried on the next instead of being abandoned with the rest of the list.
• Pay only for wines actually found. Not-found and unavailable items are flagged separately and do not count.
• Currency context follows the proxy country (France / EUR by default), so prices match the market you care about.
• Optional one-click export of every result into Notion, Linear, Airtable or Apify through MCP connectors.

Data you get

Sample shape, values are illustrative placeholders, not from a live listing.

Field	Example
wineName	2015 Sample Wine, Pomerol, France
vintage	2015
appellation	Pomerol, France
region	Pomerol
grapeVariety	Merlot
style	Red - Savoury and Classic
producer	Sample Producer
description	Short producer / wine description text.
score	97
scoreBestRating	100
criticReviewsCount	20
criticReviews	[{ "author": "Sample Critic", "score": 95, "bestRating": 100, "tastingNote": "..." }]
avgPrice	5076
avgPriceCurrency	EUR
cheapestPriceAmount	3400.00
cheapestPriceCurrency	EUR
cheapestPriceMerchant	Sample Merchant
cheapestPriceMerchantUrl	https://www.example.com/wine/0000
bottlesPerUnit	1
offers	[{ "price": 3400.0, "priceCurrency": "EUR", "merchant": "Sample Merchant", "merchantUrl": "...", "unitDescription": "Bottle (750ml)", "bottlesPerUnit": 1, "availability": "InStock" }]
offersCount	13
highestPriceAmount	4500.00
highestPriceCurrency	EUR
medianPriceAmount	3800.00
merchantCount	11
labelImageUrl	https://www.wine-searcher.com/images/labels/00/00/000000000.jpg
wineSearcherUrl	https://www.wine-searcher.com/find/sample/2015
searchLocation	France
scrapedAt	2026-01-01T00:00:00.000Z

The full record, including the complete offers array (every merchant, price and link) and the criticReviews breakdown with tasting notes, is always kept in the Apify dataset.

Specials, was-price & discount

Wine-Searcher is a price-comparison site, not a retail store. It lists merchant offers at face value and does not have a specials, deals, or clearance taxonomy, nor a was/original/list price or discount field on any offer. The price spread between merchants is not a discount: it reflects different merchants, bottle sizes, vintages, and stock conditions.

No wasPrice, savingsAmount, savingsPercent, isOnSpecial, or specialsCategory field is emitted, because none exists on the site. Wine-Searcher's "best value" and critic scores are ratings, not discounts.

Instead, the actor ships merchant-price-spread statistics computed from the already-fetched offers list:

Field	Description
cheapestPriceAmount	Lowest offer price exactly as a merchant lists it (long-standing field, meaning unchanged).
highestPriceAmount	Highest price per standard 750ml bottle (see the basis note below).
highestPriceCurrency	Currency of that highest per-bottle price.
medianPriceAmount	Median price per standard 750ml bottle, same basis as highestPriceAmount.
merchantCount	Number of distinct merchants offering this wine.
offersCount	Total number of merchant offers (already shipped).

Basis note (important). A wine's offers mix formats: single bottles, magnums, half bottles, and multi-bottle cases, and a case offer's listed price is the case total. Comparing those numbers at face value is meaningless, so highestPriceAmount and medianPriceAmount are normalised to one standard 750ml bottle: a case is divided by its bottle count (a case of 6 listed at 30,204 counts as 5,034 per bottle, not 30,204), and non-750ml formats such as magnums are left out of the comparison entirely. These two fields are therefore a like-for-like per-bottle comparison, not necessarily a price you can click and buy. If a wine has no standard-bottle offer at all, both are null. cheapestPriceAmount is untouched and keeps its original meaning (the lowest listed offer price).

These are price-spread fields for comparison across merchants, not discount fields. They are populated when fetchOffers is on and null otherwise.

How to use

Look up by wine name (auto-detected):

{
  "inputType": "auto",
  "wineNames": ["Petrus 2015", "Dom Perignon 2013"],
  "maxItems": 20
}

Look up by LWIN code:

{
  "inputType": "lwins",
  "lwins": ["11316442021", "11084042019"]
}

Look up by Wine-Searcher URL, fastest mode (no offer extraction):

{
  "inputType": "urls",
  "urls": ["https://www.wine-searcher.com/find/sassicaia/2019"],
  "fetchOffers": false
}

Pin a different currency market (United Kingdom / GBP):

{
  "inputType": "auto",
  "wineNames": ["Sassicaia 2019"],
  "proxy": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"], "apifyProxyCountry": "GB" }
}

Resume & recurring tracking

Resume from a previous run (resumeFromRunId) is a one-off delta: give it a previous run or dataset ID and this run skips wines already collected there, looking up only the rest of your list. Useful for finishing a large list across several runs without re-paying for wines you already have.

Incremental mode (incrementalMode) is for a wine list you run on a schedule (daily/weekly) to track changes over time - critic score moved, price shifted, a merchant added or dropped an offer. Every wine is classified:

changeType	Meaning
NEW	Never seen under this State key before.
UPDATED	Something changed since last time - changedFields names which top-level fields.
UNCHANGED	Identical to last time. Suppressed (not pushed, not billed) unless emitUnchanged is on.
REAPPEARED	Was EXPIRED, now found again.
EXPIRED	Tracked before, not found this run. Only reported when emitExpired is on AND this run completed a full scan of your list (no Resume, no cap, no connection ceiling cut it short).

State is kept in a separate key-value store, keyed by stateKey (or an auto-derived hash of your exact wine list when left empty) - it never re-visits wine-searcher.com, so checking state costs no extra proxy/solver spend. A page-order-only difference in offers/criticReviews never counts as a change (see the code comments in src/incremental.py for what is and isn't proven about field volatility here).

⚠️ The auto-derived key is built from your wine list itself, so editing the list (adding or removing a wine) also changes the auto key and starts a fresh baseline - convenient if you always re-run the identical list, but it means emitExpired can never report a wine you removed as EXPIRED under the auto key. If you intend to grow/shrink your tracked list over time and still want removed wines reported, set your own fixed stateKey so every run shares one baseline regardless of list edits.

{
  "inputType": "auto",
  "wineNames": ["Petrus 2015", "Opus One 2019"],
  "incrementalMode": true,
  "emitExpired": true
}

Long runs (large lists) also survive a platform migration or a manual Resurrect automatically: progress is checkpointed periodically, so a resumed attempt of the same run does not re-look-up (or re-charge for) wines it already extracted.

Send results into your apps (MCP connectors)

Optionally pipe every result into the tools you already use, no code required:

1. Authorize a connector under Apify, Settings, API & Integrations (Notion, Linear, Airtable or Apify).
2. Select it in the mcpConnectors input. For Notion, also set notionParentPageUrl.
3. Run the actor. Results are written to the connector after the dataset is saved.

The connector receives a condensed, human-readable summary per wine (a title plus the key fields). The complete record always stays in the Apify dataset. Leaving mcpConnectors empty skips this entirely and never changes the dataset output.

Input parameters

Parameter	Type	Default	Description
inputType	string	auto	auto, wineNames, urls, or lwins. Auto-detect routes each item by its shape.
wineNames	array	(none)	Wine names, ideally with vintage year. Up to 1000.
urls	array	(none)	Wine-Searcher find URLs. Up to 1000.
lwins	array	(none)	LWIN codes (7 / 11 / 16 / 18 digit). Up to 1000.
fetchOffers	boolean	false	Also extract the cheapest offer, the full offers list and popularity. Slightly slower per wine.
fetchReviews	boolean	false	Also collect the per-critic review breakdown (reviewer name and score).
concurrency	integer	3	Deprecated and ignored. Lookup parallelism is managed automatically.
maxItems	integer	20	The single cap: process at most this many wines. 0 means no limit.
proxy	object	Residential FR	Residential proxy required. Country sets the price-currency context.
mcpConnectors	array	(none)	Optional MCP connectors to export results to.
notionParentPageUrl	string	(none)	Parent Notion page for the Notion export.
maxNotifyListings	integer	50	Cap on items written to each connector per run.

Output example

Sample shape, values are illustrative placeholders, not from a live listing.

{
  "inputValue": "Petrus 2015",
  "inputType": "wineNames",
  "wineSearcherUrl": "https://www.wine-searcher.com/find/sample/2015",
  "wineName": "2015 Sample Wine, Pomerol, France",
  "vintage": 2015,
  "appellation": "Pomerol, France",
  "region": "Pomerol",
  "grapeVariety": "Merlot",
  "style": "Red - Savoury and Classic",
  "producer": "Sample Producer",
  "description": "Short producer / wine description text.",
  "score": 97,
  "scoreBestRating": 100,
  "criticReviewsCount": 20,
  "criticReviews": [
    { "author": "Sample Critic", "score": 95, "bestRating": 100, "tastingNote": "Tasted: Jan 2018. ..." }
  ],
  "avgPrice": 5076,
  "avgPriceCurrency": "EUR",
  "cheapestPriceAmount": 3400.0,
  "cheapestPriceCurrency": "EUR",
  "cheapestPriceMerchant": "Sample Merchant",
  "cheapestPriceMerchantUrl": "https://www.example.com/wine/0000",
  "bottlesPerUnit": 1,
  "offers": [
    { "price": 3400.0, "priceCurrency": "EUR", "merchant": "Sample Merchant", "merchantUrl": "https://www.example.com/wine/0000", "unitDescription": "Bottle (750ml)", "bottlesPerUnit": 1, "availability": "InStock" }
  ],
  "offersCount": 13,
  "highestPriceAmount": 4500.0,
  "highestPriceCurrency": "EUR",
  "medianPriceAmount": 3800.0,
  "merchantCount": 11,
  "labelImageUrl": "https://www.wine-searcher.com/images/labels/00/00/000000000.jpg",
  "searchLocation": "France",
  "scrapedAt": "2026-01-01T00:00:00.000Z"
}

Long lists

There is no separate setting for this: set maxItems to the size of your list and the actor sizes itself. It reads a batch of wines per connection, then opens a fresh one and carries on, running up to three connections at once once it has confirmed the site is reachable. A run that genuinely cannot reach the site still stops within a few seconds and reports a connection problem rather than an empty result, so a total block never looks like "your list matched nothing".

Plan requirement

Wine-Searcher only accepts residential connections, so this actor needs Apify Residential proxy (Starter plan or higher). On a free plan without residential access the run will return no data. The default pins France for EUR pricing; change the proxy country to switch the price-currency market. Critic scores and average price are always included; the full offers list and popularity rank are included when you enable offer fetching, and the per-critic review breakdown with tasting notes is included when you enable review collection.