TagRadar uses Steam tags to surface YouTubers. The same pattern fits SaaS, books, apps, brands. This tutorial walks the niche-agnostic version.
Prerequisites
- Python 3.10+
- Scavio API key
Walkthrough
Step 1: Define niche tags
5-15 specific tags or topics.
Python
TAGS = ['async python', 'pydantic', 'fastapi', 'ai agent dev']Step 2: YouTube search per tag
Same Scavio endpoint as TagRadar's pattern.
Python
import requests, os
API_KEY = os.environ['SCAVIO_API_KEY']
def yt(tag):
return requests.post('https://api.scavio.dev/api/v1/youtube/search',
headers={'x-api-key': API_KEY},
json={'query': f'{tag} tutorial 2026'}).json()Step 3: Aggregate by channel
Count tag overlap per channel.
Python
from collections import Counter
channels = Counter()
for tag in TAGS:
for v in yt(tag).get('videos', []):
channels[(v['channel'], v.get('channel_url'))] += 1Step 4: Filter by subscriber band
Mid-band creators (1K-100K subs) are most outreach-friendly.
Python
BAND = (1000, 100000)
filtered = {c: cnt for c, cnt in channels.items() if BAND[0] <= int(c[1].split('subscribers')[0].split()[-1].replace('K','000').replace('M','000000') or 0) <= BAND[1]}Step 5: Rank by tag overlap
Higher overlap = better fit.
Python
ranked = sorted(filtered.items(), key=lambda x: -x[1])Python Example
Python
# 5 tags × 1 query = 5 credits ≈ $0.022.JavaScript Example
JavaScript
// Same pattern in TS.Expected Output
JSON
Ranked list of niche-relevant creators with subscriber bands. Pair with outreach automation downstream.