On this page
Quick answer: Pick 5 to 10 creators in your niche, country and language. Find their outliers, meaning videos that beat that creator's own median views, not just their biggest hits. Break each one down into hook, format, length and payoff, look for a pattern across at least three, then make your own version with your angle, words and footage. TikTok's copyright policy protects how a video is expressed, not the underlying idea.
Search "TikTok competitor analysis" and you'll get guides written for brand managers. Share of voice. Campaign benchmarks. Engagement rate dashboards for five rival skincare companies. Useful if you run social for a bank. Not much use if you're one person in Multan making cooking videos on your phone.
As a creator, you want to answer two smaller questions. What's working right now for people like me? And how do I use that without becoming the account that just copies everyone?
This guide is a method for both. It's built around one idea most brand guides skip: you learn the most from videos that did unusually well for that specific creator, not from whoever has the most followers. And it draws the copying line where TikTok draws it, with links to TikTok's own text so you're not relying on our opinion.
Where TikTok draws the line
Let's start here, because it changes how you study everything else.
TikTok's copyright policy says, in general terms, that copyright protects the original expression of an idea and not the underlying ideas or facts (TikTok copyright policy, read 12 Nov 2026). The specific way a video is made, its footage, its script, its recording, belongs to whoever made it. The idea of "a budget meal under 500 rupees" does not belong to anyone.
That's TikTok's general statement, not legal advice, and copyright law differs from country to country. But it's the right starting principle.
Then there's distribution. TikTok's Community Guidelines say content is ineligible for the For You feed if it includes unoriginal or reused material without anything new, and that you should only post content you created or have the right to share (TikTok Community Guidelines, Integrity and Authenticity, effective 24 Sep 2026). So even if a copy isn't taken down, it may simply not be shown to anyone new. We covered that rule in depth in our guide to unoriginal content and faceless accounts.
Put together, here's a practical line:
| Fine to learn from | Not yours to take |
|---|---|
| The topic or question the video answers | Their footage, clips or photos |
| The format (list, tutorial, before and after, story) | Their script, word for word or lightly reworded |
| The hook type (problem, promise, contradiction) | Their exact hook line |
| Length and pacing | Their voice, face or likeness |
| Where the payoff lands | Their on-screen text and cover design |
| What commenters asked for | Their watermark, logo or brand |
| How often they post and when | Their personal story told as yours |
If you only take things from the left column and add your own angle, you're doing research. If anything from the right column ends up in your video, you're copying.
Step 1: choose who to study
Pick 5 to 10 creators. More than that and you'll stop actually watching their videos and start skimming.
Choose them on three filters.
Same niche, narrowly. "Food" is too wide. "Budget home cooking for students" is right. You want people whose viewers could be your viewers.
Same country and language. This is the filter brand guides never mention, and it matters most for creators outside the US. A format that works for an American fitness creator might fall flat in Urdu, because the jokes, the prices, the references and even the pace of speech are different. Study people who talk to your audience in your audience's language. If you post in Roman Urdu for Pakistan, a Hindi creator in Delhi might still be useful. A Texan probably isn't.
Mixed sizes. Two or three accounts bigger than you, three or four about your size, and a couple smaller ones that are growing fast. Big accounts show you what the niche rewards. Similar-sized ones show you what's possible from where you are. Fast-growing small ones show you what's new.
Where do you find them? Search your niche keywords inside TikTok in your language, look at who appears under the videos you'd like to have made, and check who the bigger creators duet or reply to. If you use Discover, filter by country and language and note which creators keep showing up in the results.
Step 2: find their outliers, not their top videos
Here's the most important step, and where most people go wrong.
Don't study a creator's most viewed video. Study the videos that did unusually well for them. A creator who always gets 2 million views and posts a video with 2 million views has told you nothing. A creator who usually gets 20,000 and posts one with 180,000 has told you something worked.
To measure "usually," use the median, not the average. The median is the middle value when you sort their recent videos by views. The average gets dragged around by one lucky video.
A made-up example to show the difference. Say a creator's last 30 videos all got around 20,000 views, except one that hit 900,000.
- Average: (29 x 20,000 + 900,000) / 30 = about 49,000.
- Median: 20,000.
Use the average and a new video with 100,000 views looks like 2x. Use the median and it's 5x, which is what it really is compared with a normal day for that account.
So the outlier multiplier is: video views divided by the creator's median views over their recent posts. Anything around 3x or above deserves a look. Above 10x is rare and worth a proper breakdown.
You can do this by hand: open the profile, write down views for the last 20 or 30 videos in a sheet, sort, and take the middle one. It takes about ten minutes per creator. Discover does this for you and shows the formula on hover, like "180K views / 20K creator median, 30 posts", plus when it was computed, so you know how fresh the number is. We explained the full search method in how to find viral TikTok videos in your niche.
Step 3: break down one outlier
Watch the outlier three times. Once as a viewer, once with the sound off, once with a notebook.
Write down:
- Hook: the first line, word for word for your notes only, and the second it ends. What type is it: a problem, a promise or a contradiction?
- Format: talking head, voiceover with B-roll, screen recording, list, before and after, story.
- Length: total seconds, and where the payoff lands.
- Payoff: what the viewer gets at the end. A result, an answer, a twist?
- On screen: what text appears, when, and how big.
- Comments: the top questions and complaints. These are free ideas.
- What only they have: their face, their kitchen, their humour, their credibility. You can't borrow this part, so name it and move on.
That last line keeps you honest. If the main reason a video worked is that it's this particular person being funny, there's not much to learn except "be yourself, more."
In Breakoutly, the Why it worked panel on each video gives you the transcript, the hook quoted with its timestamp, the format and the reasons with evidence from the video. It's a faster first pass. Still watch the video yourself.
Step 4: look for the pattern across three outliers
One outlier is an anecdote. Three that share something is a pattern.
Line up your notes. Are all three lists? Do all three put a price on screen in the first two seconds? Are they all under 30 seconds, or all over a minute? Do the hooks all start with a mistake?
If you find something shared across creators, even better. "Three different creators in my niche got 4x or more with a 'what I'd buy with 1,000 rupees' format this month" is a real finding. "One creator went viral with a dance" isn't.
Write the pattern as one sentence. That sentence is your brief.
Saving the source videos into a board in Saved, with a note on each about the hook or angle you want to try, means you can come back to the pattern next week without searching again.
Step 5: the transfer test
Now make it yours. Before you write a word, decide what you'll change. Aim for at least three of these:
- Angle. Their video is "cheap meals for students." Yours is "cheap meals when you only have one burner," because that's your actual kitchen.
- Example. Use your own city, prices, products and mistakes. A Karachi price list isn't a Lahore price list.
- Language and voice. Write it the way you talk. If they speak English and you speak Roman Urdu, you're already making a different video, but only if the script is genuinely yours, not translated line by line.
- Proof. Show your own result. Your before and after, your receipt, your screen.
- Visual. Different setting, different shots, different cover.
Then two quick checks before you post.
Would they recognise it? If the creator watched your video, would they think "nice, similar idea" or "that's my video"? You want the first one.
What did I add? Say out loud, in one sentence, what your video gives a viewer that theirs didn't. If you can't, you're not ready to post it.
If you draft in Studio from a source video, the script lists which elements it kept from the source and what it changed, so you can see the transfer at a glance and push it further. Then rewrite it in your own words before you film.
Step 6: a 30 minute weekly routine
Competitor research works when it's small and regular. One afternoon a month of frantic scrolling doesn't stick.
Here's a routine that fits in half an hour, once a week:
- 5 minutes: check your study list for new posts. In Tracking, creators you follow show their median views, best video in the last 72 hours and outlier rate, so new breakouts stand out without you opening each profile.
- 10 minutes: pick the one or two biggest outliers and do the Step 3 breakdown.
- 10 minutes: update your pattern sentence. Is it getting stronger, or was last week's pattern a fluke?
- 5 minutes: write one video idea that passes the transfer test, and put it on your plan for the week.
Every four weeks, swap out two creators on your list who haven't posted an outlier in a month for two new ones.
A worked example
Say you make budget cooking videos in Roman Urdu from Karachi. This is a made-up scenario to show the method, not real data.
You study eight creators. Over two weeks, three of them post outliers at 4x to 7x their medians. Your notes show:
- All three show the total cost in rupees on screen in the first two seconds.
- Two of the three are under 40 seconds.
- Top comments on all three ask "how many people does this feed?"
Your pattern sentence: "Short budget recipes that open with the total price on screen, and the audience wants to know serving size."
Your version: same price-first opening, but you add serving size right next to the price, because that's what commenters kept asking. Your recipe, your kitchen, your voice, your receipt from your local shop. You haven't copied any of the three videos. You've answered the question their audience was asking that none of them answered.
That's the whole game. The best ideas from competitor research are usually sitting in the comments of their outliers.
Free method vs tools
You can do all of this for free: TikTok search, profile pages, a spreadsheet and a notebook. It's slower, and calculating medians by hand gets old, but it works and it teaches you a lot.
TikTok Creative Center is free too, but in our check on 11 November 2026 its trend country list didn't include Pakistan, India or Bangladesh, and logged-out views show very little. Useful for Gulf and Southeast Asian markets, less so for South Asia.
Paid research tools save time on the median maths, the breakdowns and the weekly checking. Whatever you use, make sure it shows how each number was calculated and how fresh it is, and that it filters by your country and language. A tool that only shows US trends will lead you to study the wrong people.
One thing no legitimate research tool should do: download other creators' videos so you can re-upload them. That's the right column of the table above, and it's exactly what TikTok's originality rules are written to stop.
Mistakes to avoid
Studying only mega accounts. Their results depend on an audience you don't have yet. Mix in creators your size.
Using averages. One viral video makes every other video look like a failure. Use medians.
Copying the trend and the script together. Using a trending format is normal. Using a trending format plus someone's exact words plus a similar setting is a copy.
Translating a script line by line. An English script turned into Urdu is still their script. Write from your notes, not from their transcript.
Saving 200 videos and deciding nothing. If your notes don't end in a pattern sentence and a video idea, it's just scrolling with extra steps.
Ignoring comments. The outlier tells you what worked. The comments tell you what's missing.
FAQ
Is it okay to copy a TikTok idea?
Learning from an idea is generally fine. TikTok's copyright policy says copyright protects how something is expressed, not the underlying idea or facts. Copying someone's footage, script, voice or on-screen design is not fine, and TikTok's guidelines keep reused content without anything new out of the For You feed. This isn't legal advice, and laws differ by country.
How do I find my competitors on TikTok?
Search your niche keywords inside TikTok in the language you post in, and note which creators keep appearing. Check who shows up under videos you wish you'd made, and who bigger creators reply to. Aim for 5 to 10 creators in your niche, country and language, with a mix of bigger, similar and smaller accounts.
What is an outlier video on TikTok?
An outlier is a video that got far more views than that creator usually gets. Work it out by dividing the video's views by the creator's median views across their recent posts. A video at 3x or more is worth studying. It tells you more than a big account's normal video, because it shows something worked better than usual.
How many competitors should I track on TikTok?
Five to ten is enough for one person. Fewer and you'll miss patterns. More and you'll stop watching closely. Review them once a week for about 30 minutes, and every month replace two creators who haven't had an outlier lately with two new ones from your niche.
Can I use the same format or trend as another creator?
Yes, formats and trends are shared across TikTok, and using one is normal. What makes it yours is the angle, example, voice, proof and visuals you bring. If you use their format plus their words plus a near-identical setting, it stops being inspiration and becomes a copy, which TikTok may keep out of the For You feed.
By Breakoutly Research
Breakoutly researches short-form video for creators outside the US. Every article cites its sources and is checked by a person before it goes live.



