Retail Site Data Collection with CAPTCHA Handling

Amazon uses image CAPTCHAs to block automated access. When you hit their anti-bot threshold, you'll see a page asking you to type characters from a distorted image. CaptchaAI's OCR solving handles these automatically.

How Amazon's CAPTCHA Works

Amazon triggers CAPTCHAs based on:

Signal	Description
Request volume	Too many requests from one IP in a short window
Missing cookies	No Amazon session cookies
Suspicious headers	Bot-like User-Agent or missing headers
IP reputation	Known datacenter or proxy IP ranges

When triggered, Amazon redirects to a page with a distorted text image and an input field. You must solve the image and submit the text to continue.

Requirements

Requirement	Details
CaptchaAI API key	From captchaai.com
Python 3.7+	With requests and beautifulsoup4
Residential proxies	Recommended for sustained scraping

Solving Amazon's Image CAPTCHA

Step 1: Detect the CAPTCHA Page

import requests
from bs4 import BeautifulSoup

session = requests.Session()
session.headers.update({
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
    "Accept-Language": "en-US,en;q=0.9",
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"
})

def is_captcha_page(html):
    return "Type the characters you see in this image" in html or \
           "captcha" in html.lower()

url = "https://www.amazon.com/dp/B0EXAMPLE"
resp = session.get(url)

if is_captcha_page(resp.text):
    print("CAPTCHA detected!")
else:
    print("Page loaded successfully")

Step 2: Extract and Solve the Image

import base64

API_KEY = "YOUR_API_KEY"

def solve_amazon_captcha(session, captcha_page_html, captcha_page_url):
    soup = BeautifulSoup(captcha_page_html, "html.parser")

    # Find the CAPTCHA image
    img_tag = soup.find("img", src=lambda s: s and "captcha" in s.lower())
    if not img_tag:
        raise Exception("CAPTCHA image not found")

    img_url = img_tag["src"]

    # Download the image
    img_resp = session.get(img_url)
    img_base64 = base64.b64encode(img_resp.content).decode()

    # Submit to CaptchaAI
    submit_resp = requests.get("https://ocr.captchaai.com/in.php", params={
        "key": API_KEY,
        "method": "base64",
        "body": img_base64
    })
    task_id = submit_resp.text.split("|")[1]

    # Poll for result
    import time
    for _ in range(30):
        time.sleep(5)
        result = requests.get("https://ocr.captchaai.com/res.php", params={
            "key": API_KEY, "action": "get", "id": task_id
        })
        if result.text == "CAPCHA_NOT_READY":
            continue
        if result.text.startswith("OK|"):
            return result.text.split("|")[1]
        raise Exception(f"Solve error: {result.text}")

    raise TimeoutError("Solve timed out")

Step 3: Submit the Solution

def submit_captcha_solution(session, captcha_page_html, solution, captcha_page_url):
    soup = BeautifulSoup(captcha_page_html, "html.parser")
    form = soup.find("form")

    # Build form data
    form_data = {}
    for inp in form.find_all("input"):
        name = inp.get("name")
        if name:
            form_data[name] = inp.get("value", "")

    # Set the CAPTCHA answer
    form_data["field-keywords"] = solution

    # Submit
    action = form.get("action", captcha_page_url)
    if action.startswith("/"):
        from urllib.parse import urljoin
        action = urljoin(captcha_page_url, action)

    resp = session.post(action, data=form_data)
    return resp

Full Working Example

import requests
import base64
import time
from bs4 import BeautifulSoup

API_KEY = "YOUR_API_KEY"

def scrape_amazon_product(url):
    session = requests.Session()
    session.headers.update({
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
        "Accept-Language": "en-US,en;q=0.9"
    })

    resp = session.get(url)

    # Handle CAPTCHA if present
    if "captcha" in resp.text.lower():
        soup = BeautifulSoup(resp.text, "html.parser")
        img = soup.find("img", src=lambda s: s and "captcha" in s.lower())

        if img:
            # Download and solve
            img_data = session.get(img["src"]).content
            img_b64 = base64.b64encode(img_data).decode()

            submit = requests.get("https://ocr.captchaai.com/in.php", params={
                "key": API_KEY, "method": "base64", "body": img_b64
            })
            task_id = submit.text.split("|")[1]

            for _ in range(30):
                time.sleep(5)
                result = requests.get("https://ocr.captchaai.com/res.php", params={
                    "key": API_KEY, "action": "get", "id": task_id
                })
                if result.text == "CAPCHA_NOT_READY":
                    continue
                if result.text.startswith("OK|"):
                    solution = result.text.split("|")[1]
                    break

            # Submit solution
            form = soup.find("form")
            form_data = {inp.get("name"): inp.get("value", "")
                        for inp in form.find_all("input") if inp.get("name")}
            form_data["field-keywords"] = solution

            action = form.get("action", url)
            resp = session.post(action, data=form_data)

    # Parse product data
    soup = BeautifulSoup(resp.text, "html.parser")
    title = soup.find("span", {"id": "productTitle"})
    price = soup.find("span", class_="a-price-whole")

    return {
        "title": title.text.strip() if title else None,
        "price": price.text.strip() if price else None
    }

product = scrape_amazon_product("https://www.amazon.com/dp/B0EXAMPLE")
print(product)

Best Practices for Amazon Scraping

Use residential proxies — Amazon blocks datacenter IPs aggressively
Rotate User-Agents — Use a pool of realistic browser strings
Maintain sessions — Keep cookies across requests
Add delays — 3-10 seconds between requests
Set Accept-Language — Always include locale headers
Don't scrape logged-in pages — Product pages are accessible without login

Troubleshooting

Issue	Fix
CAPTCHA on every request	Use residential proxies; slow down request rate
CAPTCHA solution rejected	Verify image was downloaded correctly; retry
Redirect loops	Check cookie handling; use `allow_redirects=True`
Empty product data	Amazon may serve different layouts; check selectors