example.py

"""Amazon keyword research with the Amazon Search Scraper on Apify.

For each keyword: how many sponsored slots sit above the first organic result,
the median price / rating of the organic top 20, and the products with real
demand ("bought in past month") but weak review counts, i.e. open niches.

Usage:
    pip install -r requirements.txt
    set APIFY_TOKEN=<YOUR_APIFY_TOKEN>      (export on macOS/Linux)
    python example.py "standing desk" "desk lamp"
"""
import csv
import os
import statistics
import sys

from apify_client import ApifyClient

ACTOR_ID = "rel8ble/amazon-search-scraper"
DOMAIN = "amazon.com"
TOP_N = 20

token = os.environ.get("APIFY_TOKEN", "<YOUR_APIFY_TOKEN>")
keywords = sys.argv[1:] or ["standing desk", "desk lamp"]

client = ApifyClient(token)
run = client.actor(ACTOR_ID).call(run_input={
    "keywords": keywords,
    "domain": DOMAIN,
    "maxItemsPerSearch": 60,
    "includeSponsored": True,
})
items = list(client.dataset(run["defaultDatasetId"]).iterate_items())
print(f"{len(items)} products for {len(keywords)} keyword(s)")

fields = ["keyword", "position", "organicPosition", "isSponsored", "asin", "title",
          "price", "listPrice", "currency", "rating", "ratingsCount",
          "boughtLastMonth", "isBestSeller", "isAmazonChoice", "url"]
with open("amazon_search.csv", "w", newline="", encoding="utf-8") as f:
    w = csv.DictWriter(f, fieldnames=fields, extrasaction="ignore")
    w.writeheader()
    w.writerows(items)
print("Wrote amazon_search.csv")


def median(values):
    values = [v for v in values if isinstance(v, (int, float))]
    return round(statistics.median(values), 2) if values else None


for kw in keywords:
    rows = sorted((i for i in items if i.get("keyword") == kw), key=lambda i: i["position"])
    organic = [i for i in rows if not i.get("isSponsored")][:TOP_N]
    first_organic = next((i["position"] for i in rows if not i.get("isSponsored")), None)
    ads_above = sum(1 for i in rows if i.get("isSponsored") and first_organic and i["position"] < first_organic)

    print(f"\n== {kw} ==")
    print(f"sponsored slots above first organic result: {ads_above}")
    print(f"organic top {len(organic)}: median price {median(i.get('price') for i in organic)}, "
          f"median rating {median(i.get('rating') for i in organic)}, "
          f"median ratings count {median(i.get('ratingsCount') for i in organic)}")

    # Selling well, few reviews yet: room for a newcomer.
    gaps = [i for i in rows if (i.get("boughtLastMonth") or 0) >= 1000 and (i.get("ratingsCount") or 0) < 500]
    for i in sorted(gaps, key=lambda i: -(i.get("boughtLastMonth") or 0))[:5]:
        print(f"  open niche: {i.get('boughtLastMonthText')} | {i.get('ratingsCount')} ratings | "
              f"{i.get('price')} {i.get('currency')} | {i['title'][:70]}")