Shopify Scraper & eBay Scraper: Products, Variants and Prices in One Clean Schema

Copying products from competitor stores into a spreadsheet by hand is slow, and most scrapers either break on odd price formats or return an empty file without telling you why.

Give the Actor one or more Shopify store addresses, or eBay search terms. It returns a clean product dataset: one row per variant, with title, variant, vendor, product type, tags, SKU, price, compare-at price, currency, an approximate USD price, availability, images and the product URL. Every row has the same fields, whether it came from Shopify or eBay, and when a target returns nothing, the run tells you why.

• Shopify stores: reads the store's public product feed (/products.json), 250 products per page, every variant as its own row.
• eBay search (beta): reads the first results page of each search term: title, price, image and link. eBay currently refuses many automated requests, even through a residential proxy; see Limits.
• One fixed schema: a missing value is null, never a missing column.
• Honest prices: US and EU number formats and currency symbols are parsed; unclear values stay null instead of being guessed.
• Never silently empty: blocked, disabled or changed pages are reported with a status and a reason.
• Pay only for records: empty or blocked results are not billed.

What it costs: $0.005 per product row, so 100 rows cost about $0.50. Details in Pricing.

This is a deterministic scraper. There is no AI in it: what you see in the dataset is what was parsed from the source.

---

At a glance

Input	Shopify store addresses (storeUrls) or eBay search terms (searchQueries)
Output	Product dataset (one row per variant) + RUN_SUMMARY with per-target status and a data health report
Shopify fields	Product and variant IDs, title, variant title, handle, vendor, product type, tags, SKU, price, compare-at price, currency, USD price, availability, images, URL
eBay fields	Title, price, currency, USD price, image, URL, and condition when the results page shows it
Limits	maxItems rows in total (default 1,000), split evenly across targets
Statuses	OK, EMPTY_VALID, BLOCKED, ENDPOINT_DISABLED, RATE_LIMITED, STALE_SELECTOR, HTTP_ERROR
Pricing	Per product row; see Pricing
Runs on	Apify Console, API, schedules, n8n, Make, Zapier, AI agents via MCP

---

Quick start

Shopify store

1. Keep Platform on Shopify stores.
2. Enter one or more store addresses in Shopify store addresses, for example allbirds.com or https://www.allbirds.com.
3. Set Store currency (for example USD). Shopify's feed does not include a currency symbol, so this fills the currency and price_usd columns.
4. Set Maximum number of records. For a first test, 50-100 is enough.
5. Click Start and open the Dataset tab.

eBay search

1. Set Platform to eBay search.
2. Enter search terms in eBay search terms, for example mechanical keyboard.
3. Keep Use a residential proxy on, then click Start.
4. If the status is BLOCKED, eBay refused the request even after three attempts from different proxy addresses. Only the $0.0001 start fee is charged; try again later. For dependable results, use Shopify stores.

What the output looks like

source	title	variant_title	price	currency	price_usd	available	url
shopify	Wool Runner	9 / Natural Black	98.00	USD	98.00	true	https://store.example/products/wool-runner
shopify	Wool Runner	10 / Natural Black	98.00	USD	98.00	false	https://store.example/products/wool-runner
ebay	Mechanical Keyboard 87 Keys Hot Swap	null	45.99	USD	45.99	null	https://www.ebay.com/itm/...

Illustrative rows; your run returns what the store or search page actually lists.

---

Why use it

Many product scrapers work on the happy path and fail quietly on everything else. This Actor is built around four promises:

1. The schema never changes. All rows carry the same 21 fields. A value the source does not provide is null, so your spreadsheet, database or script never meets a missing column.
2. One row per variant. Sizes and colors each get their own row with their own price, SKU and availability, which is what price comparison and catalog work need.
3. Prices are parsed, not guessed. $1,299.00, 1.299,00 €, 1299 ₺ and bare numbers become a decimal amount and an ISO currency code. When a value is ambiguous, the amount stays null.
4. Empty results come with a reason. If a store disabled its product feed, eBay showed a bot check, or the page layout changed, the run says so with a status and a short explanation instead of returning an empty file.

Typical scraper vs. this Actor

	Typical product scraper	This Actor
Same columns for every row	Varies	Always; missing = null
Variants	Varies	Every variant as a row
EU price formats	Varies	Parsed; unclear stays null
Empty result	Often just an empty file	Status + reason per target
Billing on empty or blocked pages	Varies	Not billed

---

What you get

Dataset: one row per product variant

Field	Shopify	eBay
source	shopify	ebay
product_id / variant_id	Shopify IDs	null
title / variant_title	Product and variant title	Listing title / null
handle	Product handle	null
vendor / product_type	From the store	null
tags	List of tags	null
sku	Variant SKU	null
price / compare_at_price	Variant price and compare-at price	Listed price / null
currency	From Store currency	From the price symbol
price_usd	Approximate USD value	Approximate USD value
available	true / false	null
image / images	First image / all images	Listing image
url	Product page	Listing link
condition	null	Condition text when the results card shows it, otherwise null
seller / shipping	null	null in search mode

How prices are parsed

Source value	price	currency
"98.00" with Store currency USD	98.00	USD
"$1,299.00"	1299.00	USD
"1.299,00 €"	1299.00	EUR
"1299 ₺"	1299	TRY
"98.00" without Store currency	98.00	null
Text with no number	null	null

Key-value store: RUN_SUMMARY

Written when the run returns records:

• total: number of rows in the dataset.
• statuses: for each store or search term, the status, whether a retry is advised, and the reason.
• data_health: for title, price, currency, availability, image, SKU and URL, how many rows have a value and the ratio; missing_warnings lists title, price or URL gaps.

Example:

{
  "total": 312,
  "statuses": [
    {"target": "allbirds.com", "status": "OK", "retry_advised": false, "reason": "250 record(s) parsed."}
  ],
  "data_health": {
    "total_records": 312,
    "field_coverage": {"sku": {"filled": 298, "ratio": 0.96}, "price": {"filled": 312, "ratio": 1.0}},
    "missing_warnings": []
  }
}

Shortened example; the real block lists every checked field.

When nothing is returned, the dataset holds a single _summary row with the statuses and a note, and no record is billed.

---

Who it is for

• Dropshippers and e-commerce founders researching what a competitor sells and at what price.
• Pricing analysts comparing prices and discounts (compare-at price) across stores.
• Shopify agencies and developers auditing or migrating a client's catalog.
• Market researchers mapping vendors, product types and tags in a niche.
• eBay resellers checking what is listed for a search term and at which prices.
• Data teams feeding product data into a database or BI tool with a stable schema.

---

Use cases

Competitor catalog research

Run the Actor on the three to five stores you compete with. Filter by product_type and tags to see how they structure their range, and sort by price to find their entry and premium products.

Price and discount tracking

Schedule the same input daily or weekly and store each run's dataset. Comparing price and compare_at_price between runs shows new discounts and price changes. The Actor gives you clean snapshots; the comparison happens in your sheet or database.

Stock and variant availability

available is reported per variant, so you can see which sizes or colors are sold out in a competitor's store.

Catalog audits for agencies

Pull a client's catalog and check the data_health block: missing SKUs, images or prices show up as coverage below 100%.

eBay market snapshot

Search a product term and see the titles and prices on the first results page. Useful for a quick price range before buying stock or listing an item.

Brand and wholesale research

Before approaching a brand as a retailer or wholesaler, pull its direct-to-consumer store. The vendor, product type, tag and price columns show the full range, the price points and which lines are discounted, so you arrive at the conversation prepared.

Product data for your own tools

Because every row has the same fields, the dataset loads directly into Google Sheets, Airtable, a SQL table or a script without cleanup.

---

Worked example: pricing a new product line

A small shoe brand plans a new wool sneaker and wants to know how competitors price similar products.

1. Run on four competitor stores with Store currency set to USD and Maximum number of records at 800 (200 per store).
2. Filter rows by product_type or tags for sneakers.
3. Check the price range: the lowest and highest price, and which variants have a compare_at_price (a discount).
4. Check availability: sizes marked available: false across several stores hint at high demand.
5. Schedule a weekly run and keep each week's export to follow discounts and restocks.

---

How it works

1. Proxy. When Use a residential proxy is on, an Apify residential proxy is used. If a proxy cannot be set up, the run continues without one and logs a warning.
2. Shopify. For each store, the Actor requests /products.json 250 products at a time and moves to the next page until the catalog ends or the store's share of maxItems is reached.
3. eBay. For each search term, the Actor loads the first eBay search results page and reads the listing cards. Placeholder cards such as "Shop on eBay" are dropped.
4. Normalize. Every product and variant is mapped to the same 21 fields; prices are parsed into amount and currency.
5. Check. Each request gets a status: OK, or a named reason why nothing came back.
6. Deduplicate. The same product variant is kept only once.
7. Deliver. Rows go to the dataset and each row is billed as it is written; RUN_SUMMARY holds statuses and data health.

---

Input

Field	Description
platform	shopify (default) or ebay
storeUrls	Shopify store addresses, with or without https://
searchQueries	eBay search terms
maxItems	Maximum rows in total, one row per variant. Default 1,000. Split evenly across stores or search terms.
storeCurrency	Shopify store currency, for example USD, EUR, GBP, TRY. Recommended; see Tips.
useResidentialProxy	Use an Apify residential proxy. Default on.

Example: two Shopify stores

{
  "platform": "shopify",
  "storeUrls": ["allbirds.com", "https://www.example-store.com"],
  "storeCurrency": "USD",
  "maxItems": 400
}

Example: eBay search

{
  "platform": "ebay",
  "searchQueries": ["mechanical keyboard", "wireless mouse"],
  "useResidentialProxy": true
}

---

Pricing

This Actor charges per event. Current prices are on the Pricing tab; at the time of writing:

Event	When	Price
actor-start	Once per run	$0.0001
item-scraped	Per product row written to the dataset	$0.005

Rows in one run	Total (approx.)
100	$0.50
500	$2.50
1,000 (the default maximum)	$5.00
5,000	$25.00

A row is one variant, so a store with 200 products and 4 variants each can produce 800 rows. Set Maximum number of records to cap the cost of a run. When a target is blocked or returns nothing, no record is billed.

---

Integrations

API

curl -X POST "https://api.apify.com/v2/acts/enezli~shopify-ebay-product-research/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"platform":"shopify","storeUrls":["allbirds.com"],"storeCurrency":"USD","maxItems":100}'

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")
run = client.actor("enezli/shopify-ebay-product-research").call(
    run_input={"platform": "shopify", "storeUrls": ["allbirds.com"],
               "storeCurrency": "USD", "maxItems": 100})
for row in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(row["title"], row["variant_title"], row["price"], row["available"])

Schedules

Save the input as a task and add a daily or weekly schedule to build a price and stock history.

n8n, Make, Zapier

Run Actor, then read the dataset items and append them to Google Sheets, Airtable or a database. Read RUN_SUMMARY to alert when a store's status is not OK.

AI agents (MCP)

Available through the Apify MCP server, so an assistant can pull a store's catalog on request and answer questions about its prices and ranges.

---

Workflow recipes

Weekly competitor price sheet. One task with your competitors' stores and Store currency set, scheduled every Monday. An n8n or Make flow appends the rows to a Google Sheet with the run date, so a pivot table shows each variant's price over time.

Sold-out alert. Run daily on one store and keep only rows where available is false. Send the list of sold-out variants to Slack; restocks show up as rows that switch back to true.

Discount watch. Filter rows where compare_at_price is higher than price. The difference is the advertised discount; tracking it weekly shows when a competitor starts a sale.

Catalog migration check. Before and after moving a store to a new theme or app, pull the catalog and compare row counts and data_health. Missing images or SKUs show up at once.

---

Tips for better results

• Always set Store currency for Shopify. The Shopify feed gives prices without a currency symbol. Without this field, currency and price_usd stay null.
• Start small. Use maxItems 50-100 to check the output before pulling a whole catalog.
• Remember variants count as rows. Stores with many sizes and colors fill maxItems quickly.
• Keep the residential proxy on for eBay. eBay blocks many datacenter addresses.
• Treat price_usd as an estimate. It uses fixed built-in rates for USD, EUR, GBP, TRY, JPY, CAD and AUD, not live exchange rates. Other currencies get no USD value.
• Use exact store addresses. A store's own domain works best; redirects are followed automatically.
• Check RUN_SUMMARY after each run to see which targets returned OK and how complete the key fields are.

---

Limits

• Shopify stores that disabled their public product feed cannot be read; the status says ENDPOINT_DISABLED.
• eBay mode is in beta. In our September 2026 test runs, eBay refused the search page (HTTP 403) even through a residential proxy. The Actor retries up to three times with a new proxy address, then reports BLOCKED and bills nothing but the start fee.
• eBay mode reads only the first results page of each search term, and only the fields shown on the results page. Seller, shipping, item ID and SKU are not collected.
• eBay changes its page layout from time to time. When the page loads but no listing can be read, the status is STALE_SELECTOR and nothing is billed.
• For eBay price ranges such as "$20.00 to $35.00", the lower price is kept.
• The Actor does not compare runs. Price change tracking means storing snapshots and comparing them yourself.
• Only public product data is read. Password-protected stores are not supported.

---

Troubleshooting

Status or symptom	Meaning	What to do
ENDPOINT_DISABLED	The store returned 404 for its product feed	The store disabled the feed; it cannot be read with this Actor
BLOCKED	Access denied, or a bot check was detected	Turn on the residential proxy. If it is already on, the reason says so: the site is refusing automated access, try again later
RATE_LIMITED	Too many requests (HTTP 429)	Wait and retry; spread large jobs over several runs
STALE_SELECTOR	The page loaded but no product could be read	For eBay, the layout may have changed; for a store URL, check that it is a Shopify store
EMPTY_VALID	A valid but empty response	The store or search has no products
HTTP_ERROR	Server error or unexpected status	Retry later
currency is null	Store currency was not set	Set Store currency
Fewer rows than maxItems	The catalog or search page has fewer products, or the limit was split across targets	Normal; check RUN_SUMMARY
"'storeUrls' is empty"	Shopify was selected without a store	Add at least one store address

---

FAQ

Does it work with any Shopify store? With stores that keep the public /products.json feed enabled, which is the default for Shopify stores. If a store disabled it, you get the ENDPOINT_DISABLED status and no charge for records.

Why one row per variant? Because price, SKU and availability belong to the variant. One row per variant keeps each price correct and makes filtering by size or color easy.

Can I scrape several stores in one run? Yes. Add several addresses; maxItems is split evenly between them and the rows are combined in one dataset.

Do I need a Shopify account or API key? No. The Actor reads the public product feed that Shopify stores publish for visitors. No login, app install or store permission is needed.

Which eBay site is searched? eBay.com. Prices are read as shown on that site, usually in US dollars.

What data does it collect? Only public product information: titles, variants, prices, availability, images, tags and links. No customer or order data is read, and nothing behind a login is accessed.

Is there AI in this Actor? No. It parses the data the source provides. No values are generated.

How accurate is price_usd? It uses fixed conversion rates built into the Actor, so treat it as an approximate comparison value. The original price and currency are always kept.

Does it track price changes? It produces a clean snapshot each run. Schedule it and compare the snapshots in your sheet or database.

Am I charged when a store is blocked? Only the start fee ($0.0001). Records are billed only when they are written to the dataset.

What happens to duplicate products? The same product variant (same source, product ID and variant ID) is kept once. Rows without IDs, such as eBay results, are de-duplicated by link and title.

Can I use the output in a spreadsheet? Yes. Export the dataset as CSV or Excel from the Storage tab, or read it through the API. The columns are the same in every run.

Can I get more than the first eBay results page? Not in this version. Use several specific search terms to cover more listings.

---

Trademarks

This Actor is not affiliated with, endorsed by, or connected to Shopify or eBay. All product names are trademarks of their respective owners.