YouTube Trends Analyzer pulls real, current videos from the official YouTube
Data API — either matched to a niche keyword, or straight from YouTube's
own per-country Trending chart — ranks them by view velocity (views per
day since publish, the real signal of what's trending right now rather than
old evergreen content), and groups their titles into semantic topic clusters.
No AI-generated topic guesses, just what's actually performing on YouTube today.
FAQ
How the trend analysis and clustering work.
How does this tool find trending topics in a niche?+
It queries the official YouTube Data API for videos matching your niche keyword, published within your chosen time window, ordered by view count. Each video's view velocity (views per day since publish) is computed, then titles are grouped into semantic topic clusters using sentence embeddings — so you see trending topics, not just a flat list of videos.
What is "view velocity" and why use it instead of raw views?+
View velocity is views divided by days since the video was published. A 2-day-old video with 50,000 views is trending harder than a 2-year-old video with 500,000 views — raw view count rewards old evergreen content, velocity surfaces what's actually gaining traction right now.
Does this use real YouTube data or AI-generated guesses?+
Real data only. Every video, view count, like count, and publish date comes directly from the YouTube Data API v3 — nothing here is generated or estimated by a language model.
How much does it cost to run an analysis?+
Nothing to you — it runs on the free YouTube Data API quota (10,000 units/day). Each analysis costs roughly 105 units (one search call + one stats call), so the free tier supports about 90 analyses per day.
What does the semantic topic clustering do?+
It embeds every video title with a multilingual sentence-transformer model and groups titles that are close in meaning, even if they don't share exact words. Clusters are ranked by combined view velocity, so the first cluster is the hottest topic in the niche right now.
What regions can I analyze?+
Any region the YouTube Data API supports, picked from the Region dropdown (US, UA, GB, DE, and others). Note that regionCode mostly affects which videos are eligible/available in that market, not a strict geo-filter of who uploaded them — a Ukrainian creator's video can still surface under a US search if it's popular there too.
What time windows are available, and which should I pick?+
7, 14, 30, or 90 days. Shorter windows (7-14 days) surface what's genuinely fresh and rising right now; longer windows (30-90 days) pull in more videos and more channels, but also let older "slow risers" outrank this week's actual breakout topics. Start at 30 days and narrow down if results feel stale.
What is "engagement rate" and how is it calculated?+
(likes + comments) ÷ views, shown as a percentage. It's a rough signal of how much a video resonates with the people who actually watch it, independent of raw view count — a smaller video with high engagement can indicate a topic worth covering even if its view velocity is modest.
How many videos does one analysis look at?+
Up to 50 videos per niche query, pulled from a single YouTube search.list call and then enriched with statistics in a second call. That's enough to see the shape of a niche's current top content without burning excessive API quota.
Can I filter results by language?+
Yes, optionally. Leave it blank to search regardless of language, or enter a two-letter code (en, uk, es, etc.) to bias results toward that language via YouTube's relevanceLanguage parameter. It's a relevance hint, not a hard filter, so an occasional other-language video can still appear.
Why do some videos show 0 likes or missing stats?+
Since 2021, YouTube lets creators hide their public like count — the API then reports 0 (or omits the field) for those videos. That's expected behavior from YouTube itself, not a bug in this tool; view count and comment count are unaffected.
How is this different from YouTube's own Trending tab?+
YouTube's Trending tab is platform-wide, curated per country and content category, and doesn't let you target a specific niche keyword. This tool searches and ranks videos for the exact niche you type in, which is far more useful for content research than a general trending feed.
Can I use this for content ideas, not just research?+
Yes — that's the main use case. Skim the topic clusters for angles that are currently working in your niche (formats, comparisons, specific products or questions), then use them as a starting point rather than copying titles outright.
Can I export the results or use them programmatically?+
Yes. Every finished analysis has a "download CSV" link with the topic clusters and full video stats, plus a JSON view. You can also skip the UI and call POST /api/analyze with a niche/region/days, then poll GET /api/jobs/{id} for the clustered result.
Does this work for very small or ultra-specific niches?+
It depends on how much content actually exists on YouTube for that niche — this tool only surfaces real videos, it doesn't invent topics when data is thin. For a very narrow niche you may get fewer results or looser clusters; broadening the time window or the keyword usually helps.
Where do the "popular YouTube search queries" come from?+
From YouTube's own search-box autocomplete (an unofficial endpoint, not the Data API), expanded across the alphabet and common modifiers like "tutorial", "review", or "vs". These are real query completions YouTube itself suggests, but YouTube doesn't publish actual search-volume numbers anywhere, so this list isn't ranked by popularity — treat it as a set of real phrasings worth considering, not a ranked chart.
YouTube's audience growth
For context on the scale this data comes from: YouTube's monthly active users (MAU)
over time. 2013, 2017 and 2019 are YouTube's own officially announced milestones;
2021–2024 are third-party estimates aggregated from public reporting
(DataReportal, Business of Apps), since YouTube doesn't publish MAU every year.
Monthly active users, worldwide. Sources: YouTube/Google official announcements (2013, 2017, 2019) ·
DataReportal & Business of Apps aggregated estimates (2021–2024).