Scrape Twitter followers with Python

You don’t need a headless browser or a logged-in X account to get a follower list. The RelayX API returns followers as JSON, one page at a time. This guide exports a whole list to CSV.

You want	Endpoint	Returns
Followers with full profiles	GET /twitter/user/followers	followers[] (user objects)
Who an account follows	GET /twitter/user/followings	followings[]
Just the IDs, fast and cheap	GET /twitter/user/followers_ids	ids[]

Export followers to CSV

import csv, requests

API = "https://api.relayxapi.com"
HEADERS = {"X-API-Key": "YOUR_KEY"}

def followers(user, max_pages=50):
    cursor = None
    for _ in range(max_pages):
        params = {"userName": user, **({"cursor": cursor} if cursor else {})}
        r = requests.get(f"{API}/twitter/user/followers", params=params, headers=HEADERS)
        r.raise_for_status()
        body = r.json()
        yield from body.get("followers", [])
        if not body.get("has_next_page"):
            break
        cursor = body["next_cursor"]

with open("followers.csv", "w", newline="", encoding="utf-8") as f:
    w = csv.writer(f)
    w.writerow(["id", "userName", "name", "followers", "following", "verified", "createdAt"])
    for u in followers("nasa", max_pages=5):
        w.writerow([u["id"], u["userName"], u["name"], u["followers"], u["following"],
                    u["isBlueVerified"], u["createdAt"]])

max_pages is your spending cap: every page costs credits, so stop as soon as you have what you need. Swap followers for followings (response key followings) to get who an account follows.

Only need the IDs?

followers_ids returns up to 5,000 IDs per call and costs much less per item. It’s the right choice for diffing lists (who unfollowed, mutuals) or counting overlap between accounts:

def follower_ids(user):
    cursor, ids = -1, []
    while True:
        body = requests.get(f"{API}/twitter/user/followers_ids", headers=HEADERS,
                            params={"userName": user, "cursor": cursor}).json()
        ids += body.get("ids", [])
        if not body.get("has_next_page"):
            return ids
        cursor = body["next_cursor"]

a, b = set(follower_ids("nasa")), set(follower_ids("Space_Station"))
print(len(a & b), "accounts follow both")

Hydrate just the IDs you care about with GET /twitter/user/batch_info_by_ids.

What it costs

Followers are billed per user returned, and the rate drops as pages fill: 3 → 2 → 1 credits each. Follower IDs cost 2 → 1 → 0.4 credits each. With 100,000 credits = $1, 1,000 followers cost $0.01–$0.03 and 1,000 IDs cost well under a cent. See credits in practice and pricing.

Why not run a scraper?

Self-hosted X scrapers need logged-in accounts, proxies and constant fixes whenever X changes its web app. Accounts used for scraping get rate-limited or suspended. The API gives you the same public data over plain HTTPS, and keeping it working is our job. More on this: Twitter scraper vs API.