How to Scrape Amazon Reviews with Python: Ultimate Guide for 2026

Python
import requests
from bs4 import BeautifulSoup
import pandas as pd

asin = "B098FKXT8L"
custom_headers = {
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8",
    "Accept-Language": "en-US,en;q=0.9",
    "Connection": "keep-alive",
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) Gecko/20100101 Firefox/135.0"
}

def get_soup(url):
    response = requests.get(url, headers=custom_headers)
    if response.status_code != 200:
        print("Error in getting webpage")
        return None
    return BeautifulSoup(response.text, "lxml")

def extract_review(review, is_local=True):
    author_element = review.select_one(".a-profile-name")
    author = author_element.text.strip() if author_element else "Unknown"

    rating_element = review.select_one(".review-rating > span")
    rating = rating_element.text.replace("out of 5 stars", "").strip() if rating_element else "0"

    date_element = review.select_one(".review-date")
    date = date_element.text.strip() if date_element else "Unknown"

    if is_local:
        title = review.select_one(".review-title span:not([class])").text.strip() if review.select_one(".review-title span:not([class])") else ""
        content = ' '.join(review.select_one(".review-text").stripped_strings) if review.select_one(".review-text") else ""
        img_selector = ".review-image-tile"
    else:
        title = review.select_one(".review-title .cr-original-review-content").text.strip() if review.select_one(".review-title .cr-original-review-content") else ""
        content = ' '.join(review.select_one(".review-text .cr-original-review-content").stripped_strings) if review.select_one(".review-text .cr-original-review-content") else ""
        img_selector = ".linkless-review-image-tile"

    verified_element = review.select_one("span.a-size-mini")
    verified = verified_element.text.strip() if verified_element else None

    image_elements = review.select(img_selector)
    images = [img.attrs.get("data-src") for img in image_elements if img.attrs.get("data-src")] if image_elements else None

    return {
        "type": "local" if is_local else "global",
        "author": author,
        "rating": rating,
        "title": title,
        "content": content.replace("Read more", ""),
        "date": date,
        "verified": verified,
        "images": images
    }

def get_reviews(soup):
    reviews = []
    local_reviews = soup.select("#cm-cr-dp-review-list > div[data-hook='review']")
    global_reviews = soup.select("#cm-cr-global-review-list > div[data-hook='review']")
    for review in local_reviews:
        reviews.append(extract_review(review, is_local=True))
    for review in global_reviews:
        reviews.append(extract_review(review, is_local=False))
    return reviews

def main():
    search_url = f"https://www.amazon.com/dp/{asin}"
    soup = get_soup(search_url)
    if soup is None:
        return
    reviews = get_reviews(soup)
    df = pd.DataFrame(reviews)
    df.to_csv(f"reviews_{asin}.csv", index=False)
    print("Data exported to CSV.")

if __name__ == "__main__":
    main()