How to Find Your Competitors' Best-Performing Videos on YouTube (Legally, Free)
Use the YouTube Data API v3 and public channel pages to pull your competitors' top videos, view counts, and publish patterns without paying for a tool.
To find your competitors' best-performing videos on YouTube legally and for free, use the official YouTube Data API v3 to pull each channel's uploads, then sort the results by view count. The API returns public metrics for any YouTube channel: viewCount, likeCount, commentCount, publishedAt, title, description, duration, and thumbnail. You do not need the creator's permission, and you do not need to scrape anything.
This guide is for Shopify and WooCommerce merchants who want to study what is working for competing brands on YouTube Shorts and long-form without paying for a third-party intel tool. You will learn how to request an API key, which three endpoints to combine, how to spot a channel's hero videos, and how to turn that data into a weekly content signal for your own publishing calendar.
The Short Answer: Finding Competitor Hero Videos
Register a Google Cloud project, enable the YouTube Data API v3, and grab the free API key. Call search.list or channels.list to resolve a channel handle to its channel ID, call playlistItems.list on the channel's uploads playlist to get every public video, then call videos.list?part=statistics,snippet,contentDetails with up to 50 video IDs at a time to pull view counts and metadata. Sort by viewCount descending and you have their top performers.
Why This Works Without Scraping
YouTube publishes view counts, likes, and comments on every public video. The Data API v3 is the official read interface Google provides for third parties to consume this data. Reading it respects Google's terms of service, does not require a login, and does not depend on fragile web-scraping that breaks every time YouTube changes markup. The free quota is generous: 10,000 units per day, and a videos.list call costs 1 unit per 50 videos. You can audit 500 competitor videos a day without spending a cent.
What Signals Does the API Actually Return?
- viewCount — cumulative public view count on each video.
- likeCount — public like count (can be hidden by creator, usually not).
- commentCount — public comment count.
- publishedAt — timestamp of upload.
- title, description, tags — exactly what the creator wrote.
- contentDetails.duration — ISO 8601 duration; anything under PT60S is a Short.
- thumbnails — URLs for the channel's custom thumbnail choices.
What you do not get: watch-time, average view duration, retention curves, demographic data. None of the big three platforms expose that to third parties for anyone but the account owner. Anyone claiming to show you a competitor's retention curve is almost certainly scraping or estimating.
Step-By-Step: Pulling a Competitor's Top Videos
- Open Google Cloud Console. Create a new project named something like "byvano-competitor-intel".
- Enable YouTube Data API v3. Under APIs and Services, search for it and click enable.
- Create credentials. Generate an API key. Restrict it to the YouTube Data API v3 for safety.
- Resolve the channel handle. If you have @competitorhandle, call channels.list?forHandle=@competitorhandle&part=id,snippet,contentDetails and read contentDetails.relatedPlaylists.uploads. That playlist ID is the channel's full upload history.
- Walk the uploads playlist. Call playlistItems.list?playlistId=UU...&part=contentDetails&maxResults=50, paginate with pageToken until the playlist ends. Collect every videoId.
- Batch into videos.list. Call videos.list?id=ID1,ID2,...,ID50&part=statistics,snippet,contentDetails. The 50-per-call batch costs 1 quota unit total.
- Sort by viewCount. You now have the full upload history with view counts. Sort descending and the top 10 are the hero videos.
- Save a snapshot. Store what you pulled with today's date. Weekly snapshots let you detect which videos are still accruing views and which have plateaued.
How to Separate Breakouts From Steady Winners
A channel's top-10 by lifetime view count is mostly their oldest videos. That is noise; those videos had years to accumulate views. The real signal is videos that outperform the channel's own baseline. Compute the median view count of the channel's last 10 uploads. Any new upload that passes 2x that median within its first week is a breakout. This is the same rule experienced social teams apply manually, and it filters out the "this is just an older video that had time" false positive. For a deeper walkthrough see our post on the 2x baseline rule.
Which Metrics Matter Most for Ecommerce?
For an ecommerce merchant studying competitors, rank the signals in this order:
- View-to-subscriber ratio. If a 5,000-subscriber channel landed a 100,000-view video, something about that video broke out of their normal audience.
- Likes-to-views ratio. Above 4 percent is unusually engaged. Below 1 percent is passive scrolling.
- Comment volume on short-form. Shorts rarely earn comments, so a Shorts video with hundreds of comments is almost certainly over-performing.
- Days-since-upload versus views. 500,000 views in 48 hours is a breakout. 500,000 views over three years is longevity, which is useful but a different lesson.
How Often Should I Re-Pull the Data?
For active competitor watching, twice a day is plenty. A fresh video takes at least a few hours to show its true trajectory on YouTube, and the API's view count updates in near-real-time but with a few-minute delay. If you are running this manually, once a day on the morning you plan your content is enough. If you automate it as a background job, every 12 hours keeps the data fresh without burning quota.
Turning the Data Into a Weekly Content Plan
Each week, look at the five breakout videos across your tracked competitors and ask three questions for each one. What is the hook in the first three seconds? What is the structure (demonstration, before-and-after, problem-reveal, listicle)? What is the call to action, if any? Those three answers become your brief for the following week's shoots. You are not copying the video; you are studying the format that worked and adapting it to your own product. Once you have a draft script, you can schedule the output across platforms with a tool like Byvano's scheduler.
What About Instagram and TikTok?
YouTube is the easiest platform to study legally because the Data API is open, generous, and documented. Instagram exposes competitor data only through the Graph API's Business Discovery feature, which requires your own IG Business account and returns a limited slice of the target's data (follower count, recent media, no individual video views). TikTok offers almost nothing for third-party competitor intel unless you are approved for the Research API, which is a slow, academic-grade process. For 2026, YouTube Data API v3 is where realistic competitor intel starts.
Common Mistakes When Pulling Competitor Data
- Querying one video at a time. Batch 50 IDs per videos.list call to save quota.
- Trusting a single snapshot. Views change. A weekly series of snapshots beats one lifetime pull.
- Ignoring duration. A long-form video and a Short with the same view count are not the same achievement. Filter by contentDetails.duration before you draw conclusions.
- Chasing absolute view counts. A 50,000-view video on a 10,000-subscriber channel is more interesting than a 500,000-view video on a 10-million-subscriber channel.
- Scraping when the API exists. Scrapers break, violate ToS, and put your business at risk. The API is free and documented.
Keeping the Data Pipeline Small
You do not need a data warehouse to do this well. A spreadsheet with one tab per tracked channel, refreshed weekly, is enough for a store owner who tracks five to ten competitors. Each tab holds videoId, title, publishedAt, viewCount, and a column for your own notes on the hook. At the end of the quarter you will have a durable library of what worked in your niche, built on official data, with zero third-party dependency.