Scraping Online Auction Data: eBay, Sotheby's, and Christie's Price Intelligence

Why Auction Data Matters

Auction prices are the purest signal of market value — what someone actually paid, not what a seller hoped for. Whether you're flipping collectibles, investing in art, or doing market research, programmatic access to auction results is incredibly valuable.

The Challenge

Each platform structures data differently:

• eBay — completed listings with bid histories
• Sotheby's — high-end lots with estimate ranges
• Christie's — detailed provenance and hammer prices

Setup

pip install requests beautifulsoup4 pandas selenium

Auction sites are heavily protected. ScraperAPI handles rotating proxies and browser fingerprinting so you can focus on parsing.

Scraping eBay Completed Listings

import requests
from bs4 import BeautifulSoup
import pandas as pd
from datetime import datetime

SCRAPER_API = "http://api.scraperapi.com"
API_KEY = "YOUR_SCRAPERAPI_KEY"

def scrape_ebay_sold(query, pages=3):
    """Scrape completed/sold listings from eBay."""
    all_items = []

    for page in range(1, pages + 1):
        target = f"https://www.ebay.com/sch/i.html?_nkw={query}&LH_Complete=1&LH_Sold=1&_pgn={page}"
        url = f"{SCRAPER_API}?api_key={API_KEY}&url={target}"

        response = requests.get(url)
        soup = BeautifulSoup(response.text, "html.parser")

        for item in soup.select(".s-item"):
            title_el = item.select_one(".s-item__title")
            price_el = item.select_one(".s-item__price")
            date_el = item.select_one(".s-item__ended-date, .s-item__endedDate")

            if title_el and price_el:
                all_items.append({
                    "title": title_el.text.strip(),
                    "sold_price": parse_price(price_el.text),
                    "sold_date": date_el.text.strip() if date_el else None,
                    "source": "ebay"
                })

    return pd.DataFrame(all_items)

def parse_price(text):
    """Extract numeric price from text like ,234.56."""
    import re
    match = re.search(r"[\d,]+\.?\d*", text.replace(",", ""))
    return float(match.group()) if match else None

Scraping Sotheby's Auction Results

def scrape_sothebys_results(category="watches", limit=50):
    """Scrape auction results from Sothebys."""
    target = f"https://www.sothebys.com/en/results?query={category}"
    url = f"{SCRAPER_API}?api_key={API_KEY}&url={target}&render=true"

    response = requests.get(url)
    soup = BeautifulSoup(response.text, "html.parser")

    lots = []
    for lot in soup.select(".LotCard, [data-testid=lot-card]"):
        title = lot.select_one(".lot-title, h3")
        estimate = lot.select_one(".estimate")
        hammer = lot.select_one(".hammer-price, .sold-price")

        if title:
            lots.append({
                "title": title.text.strip(),
                "estimate": estimate.text.strip() if estimate else None,
                "hammer_price": parse_price(hammer.text) if hammer else None,
                "source": "sothebys"
            })

    return pd.DataFrame(lots[:limit])

Christie's Results

def scrape_christies_results(search_term):
    """Scrape auction results from Christies."""
    target = f"https://www.christies.com/en/results?keyword={search_term}&isautosuggestclick=false&saession=sold"
    url = f"{SCRAPER_API}?api_key={API_KEY}&url={target}&render=true"

    response = requests.get(url)
    soup = BeautifulSoup(response.text, "html.parser")

    results = []
    for lot in soup.select(".lot-container, .chr-lot"):
        title = lot.select_one(".lot-title, .chr-lot__title")
        price = lot.select_one(".price-realised, .chr-lot__price")
        sale = lot.select_one(".sale-title, .chr-lot__sale")

        if title and price:
            results.append({
                "title": title.text.strip(),
                "hammer_price": parse_price(price.text),
                "sale": sale.text.strip() if sale else None,
                "source": "christies"
            })

    return pd.DataFrame(results)

Cross-Platform Analysis

def price_analysis(query):
    """Aggregate and analyze prices across platforms."""
    ebay = scrape_ebay_sold(query)
    sothebys = scrape_sothebys_results(query)
    christies = scrape_christies_results(query)

    all_data = pd.concat([ebay, sothebys, christies], ignore_index=True)
    all_data = all_data.dropna(subset=["sold_price", "hammer_price"], how="all")

    # Normalize price column
    all_data["price"] = all_data["sold_price"].fillna(all_data["hammer_price"])

    print(f"\n📊 Price Analysis for {query}")
    print(f"Total results: {len(all_data)}")
    print(f"Mean price:   ${all_data[price].mean():,.2f}")
    print(f"Median price: ${all_data[price].median():,.2f}")
    print(f"Price range:  ${all_data[price].min():,.2f} - ${all_data[price].max():,.2f}")
    print(f"\nBy platform:")
    print(all_data.groupby("source")["price"].describe())

    return all_data

Proxy Strategy

Auction sites employ sophisticated bot detection. Layer your approach:

1. ScraperAPI for general scraping with auto-retry
2. ThorData residential proxies for location-specific pricing
3. ScrapeOps for monitoring scraper health and success rates

Legal Considerations

• eBay completed listings are publicly accessible
• Auction results are generally public records
• Respect rate limits and Terms of Service
• Don't scrape personal bidder information

Conclusion

Auction data scraping gives you pricing intelligence that's impossible to get any other way. Start with eBay for volume, then add premium houses for high-value categories. The combination gives you the most complete picture of what things actually sell for.