YouTube search data reveals what content performs in any niche, which creators dominate, and what gaps exist. This pipeline builds an agent that researches a topic on YouTube, ranks channels by consistency, and outputs a competitive landscape report. Each YouTube search costs $0.005 through Scavio, making a full niche analysis under $0.10.
Prerequisites
- Python 3.8+
- requests library
- A Scavio API key from scavio.dev
- Target niches or topics to research
Walkthrough
Step 1: Search YouTube for topic coverage
Query YouTube for videos on a topic and extract channel data.
Python
import os, requests, json
from collections import Counter, defaultdict
API_KEY = os.environ['SCAVIO_API_KEY']
SH = {'x-api-key': API_KEY, 'Content-Type': 'application/json'}
def youtube_search(query):
data = requests.post('https://api.scavio.dev/api/v1/search',
headers=SH, json={'query': query, 'platform': 'youtube', 'country_code': 'us'}).json()
results = data.get('organic_results', data.get('video_results', []))[:10]
return [{'title': r.get('title', ''), 'channel': r.get('channel', {}).get('name', r.get('channel_name', '')),
'views': r.get('views', 0), 'date': r.get('published_date', ''),
'link': r.get('link', '')} for r in results]
videos = youtube_search('python web scraping tutorial 2026')
print(f'Found {len(videos)} videos')
for v in videos[:5]: print(f' {v["channel"]:25} | {v["title"][:50]}')Step 2: Map the channel landscape
Search multiple angles to build a map of who covers the topic.
Python
def map_landscape(topic, angles=None):
if not angles:
angles = [f'{topic} tutorial', f'{topic} guide 2026', f'{topic} for beginners',
f'best {topic}', f'{topic} tips']
channels = defaultdict(lambda: {'videos': 0, 'total_views': 0, 'titles': []})
for angle in angles:
videos = youtube_search(angle)
for v in videos:
ch = v['channel'] or 'Unknown'
channels[ch]['videos'] += 1
channels[ch]['total_views'] += v.get('views', 0) if isinstance(v.get('views'), int) else 0
channels[ch]['titles'].append(v['title'][:60])
cost = len(angles) * 0.005
print(f'\nLandscape for "{topic}" ({len(angles)} queries, ${cost:.3f}):')
sorted_ch = sorted(channels.items(), key=lambda x: x[1]['videos'], reverse=True)
for name, data in sorted_ch[:10]:
print(f' {name:25} | {data["videos"]} videos | views: {data["total_views"]:,}')
return dict(channels), cost
channels, cost = map_landscape('web scraping')Step 3: Identify content gaps
Find subtopics with low competition or outdated coverage.
Python
def find_gaps(topic, subtopics):
gaps = []
for sub in subtopics:
query = f'{topic} {sub}'
videos = youtube_search(query)
recent = [v for v in videos if '2026' in v.get('date', '') or '2025' in v.get('date', '')]
total = len(videos)
print(f' {sub:25} | {total} results | {len(recent)} recent')
if total < 5 or len(recent) < 2:
gaps.append({'subtopic': sub, 'total': total, 'recent': len(recent),
'opportunity': 'low competition' if total < 5 else 'outdated content'})
print(f'\nGaps found: {len(gaps)}')
for g in gaps:
print(f' {g["subtopic"]}: {g["opportunity"]} ({g["total"]} total, {g["recent"]} recent)')
return gaps
subtopics = ['playwright', 'scrapy', 'beautifulsoup', 'selenium', 'httpx', 'curl_cffi']
gaps = find_gaps('web scraping', subtopics)Step 4: Generate the landscape report
Combine channel mapping and gap analysis into a report.
Python
def generate_report(topic, subtopics=None):
if not subtopics:
subtopics = ['beginner', 'advanced', 'tools', 'automation', 'api']
print(f'=== YouTube Landscape Report: {topic} ===')
channels, map_cost = map_landscape(topic)
print(f'\n--- Content Gaps ---')
gaps = find_gaps(topic, subtopics)
gap_cost = len(subtopics) * 0.005
total_cost = map_cost + gap_cost
top_channels = sorted(channels.items(), key=lambda x: x[1]['videos'], reverse=True)[:5]
report = {
'topic': topic,
'total_channels': len(channels),
'top_5': [{'name': n, 'videos': d['videos']} for n, d in top_channels],
'gaps': gaps,
'cost': total_cost
}
with open(f'yt_report_{topic.replace(" ", "_")}.json', 'w') as f:
json.dump(report, f, indent=2)
print(f'\nTotal cost: ${total_cost:.3f}. Saved report.')
return report
generate_report('web scraping')Python Example
Python
import os, requests
from collections import Counter
SH = {'x-api-key': os.environ['SCAVIO_API_KEY'], 'Content-Type': 'application/json'}
def yt_landscape(topic):
channels = Counter()
for suffix in ['tutorial', 'guide 2026', 'for beginners']:
data = requests.post('https://api.scavio.dev/api/v1/search',
headers=SH, json={'query': f'{topic} {suffix}', 'platform': 'youtube', 'country_code': 'us'}).json()
for r in data.get('organic_results', [])[:10]:
channels[r.get('channel', {}).get('name', 'Unknown')] += 1
print(f'{topic} YouTube landscape ({len(channels)} channels):')
for ch, count in channels.most_common(5):
print(f' {ch}: {count} videos')
yt_landscape('web scraping')JavaScript Example
JavaScript
const SH = { 'x-api-key': process.env.SCAVIO_API_KEY, 'Content-Type': 'application/json' };
async function ytLandscape(topic) {
const channels = {};
for (const suffix of ['tutorial', 'guide 2026', 'for beginners']) {
const data = await fetch('https://api.scavio.dev/api/v1/search', {
method: 'POST', headers: SH,
body: JSON.stringify({ query: `${topic} ${suffix}`, platform: 'youtube', country_code: 'us' })
}).then(r => r.json());
for (const r of (data.organic_results || []).slice(0, 10)) {
const ch = r.channel?.name || 'Unknown';
channels[ch] = (channels[ch] || 0) + 1;
}
}
const sorted = Object.entries(channels).sort((a,b) => b[1]-a[1]);
console.log(`${topic}: ${sorted.length} channels`);
sorted.slice(0, 5).forEach(([ch, n]) => console.log(` ${ch}: ${n} videos`));
}
ytLandscape('web scraping').catch(console.error);Expected Output
JSON
Landscape for "web scraping" (5 queries, $0.025):
Tech With Tim | 4 videos | views: 2,340,000
Corey Schafer | 3 videos | views: 1,890,000
John Watson Rooney | 3 videos | views: 456,000
NetworkChuck | 2 videos | views: 3,200,000
Fireship | 2 videos | views: 5,100,000
--- Content Gaps ---
playwright | 8 results | 3 recent
scrapy | 9 results | 2 recent
beautifulsoup | 10 results | 1 recent
httpx | 3 results | 1 recent
curl_cffi | 2 results | 0 recent
Gaps found: 2
httpx: low competition (3 total, 1 recent)
curl_cffi: low competition (2 total, 0 recent)
Total cost: $0.055. Saved report.