Competitor analysis for content creators: the creator-scale method
Competitor analysis for a creator means scoring each video a rival posts against that creator’s own usual, their median views, rather than against brand benchmarks averaged over 90 days. Track median views, outlier multiple, hit rate, views per day and saves and shares per 1,000 views, then copy the mechanic behind the outliers, not the topic.
Updated Oct 26, 2026 · based on 522 videos from 478 creators · next refresh ~Nov 2, 2026 · How we measure
Most competitor analysis guides are written for brand teams with a quarterly report to fill. A creator’s question is narrower: which videos in my niche beat their creator’s normal this month, and what did they do differently? This guide answers it with seven numbers you can work out from public profiles, a worked teardown of a beauty account and a scorecard you can copy.
The examples come from beauty TikTok and beauty Reels, but the method works in any niche. We make Followlly, so weigh our view accordingly. Everything here also works with a spreadsheet and no tool at all.
Why brand-scale competitor analysis fails creators
Brands ask how they did against other brands this quarter. Creators need to know which video beat its creator’s normal, and why.
The top-ranking competitor analysis guides are built for brand teams, and three of their habits hurt a creator:
- Averages. Sprout Social’s TikTok guide benchmarks follower growth, average views per video and average engagement rate. One viral video drags an average up, so normal videos look weak and the hit looks smaller than it was.
- Long windows. Sotrender’s Instagram guide asks for 30 days of data, ideally 90, partly so one-off viral spikes don’t distort the picture. For a creator the spike is the signal, and by day 90 the format may be crowded.
- Per-follower rates. Sotrender’s engagement benchmark, roughly 0.4–0.8%, divides by followers. Reels and TikToks reach many people who don’t follow the account, so a per-follower rate mixes up how far a video traveled with how its viewers reacted.
Influencer Marketing Hub’s TikTok guide (updated Sep 28, 2026) comes closest, listing median views, views per follower and a “hit rate” without formulas. Here is the difference on an illustrative beauty account with twelve recent videos and 85K followers:
| Brand-scale reading | Creator-scale reading | |
|---|---|---|
| A normal video | 32,500 views (average) | 20,500 views (median) |
| The 140K video | 4.3× the average | 7× the median of the other eleven: an outlier |
| The 52K video | 1.6×: a bit above average | 2.6×: a near-miss worth watching |
| How often they beat their usual | Not measured | Two of 12 videos at 2× or more (hit rate 17%) |
| Engagement | Interactions ÷ 85K followers | Saves and shares per 1,000 views, video by video |
| Time frame | A 30–90 day total | Each video on its own, once it is three days old |
Source: Sotrender (Apr 10, 2026), Sprout Social (Feb 4, 2026), Influencer Marketing Hub, checked Oct 26, 2026. The account is illustrative: every figure in the table is computed from its made-up but typical numbers.
The creator-scale metrics
All measured against the creator’s own normal, from public numbers.
| Metric | How to calculate it | What it tells you |
|---|---|---|
| Median views | The middle value of the last 10–20 videos older than three days | The creator’s usual. One viral video barely moves it. |
| Outlier multiple | Views ÷ the median of the creator’s other videos | 3× or more is an outlier; 2–3× is a near-miss. |
| Hit rate | Share of videos at 2× the median or more | A repeatable format, or one lucky video |
| Views per day | Views ÷ age in days, counting at least one day | Speed, comparable across video ages |
| Saves per 1,000 views | Saves ÷ views × 1,000 (public on TikTok only) | Whether people keep it to come back to |
| Shares per 1,000 views | Shares ÷ views × 1,000 (shown on every TikTok) | Whether people send it on |
| Engagement rate by views | (Likes + comments + shares + saves) ÷ views × 100 | How strongly the people who saw it reacted |
hit rate = videos at 2× or more ÷ videos scored
On Reels, saves are private and likes can be hidden, so use comments per 1,000 views there.
Why views beat followers as the denominator: on the illustrative account, a typical 20K-view video scores 9.2% by views and 1.8% by followers, and the 140K hit scores 9.5% and 12.2%. By followers the hit looks 6.6× as engaging, yet by views people reacted about as strongly to both. What changed is saves, from 13.0 to 20.7 per 1,000 views, and shares, from 1.5 to 3.0: people kept the hit and passed it on, the first clue to its mechanic. The engagement rate calculator shows both rates side by side.
In our beauty TikTok sample (145 videos, last 21 days), the median video earned 18.1 saves and 2.3 shares per 1,000 views, with a median engagement rate of 13.8% by views. 50% of the videos with a known follower count got more views than their creator has followers, and the median rate by followers is 7.9%.
In our beauty Reels sample (377 Reels), the median Reel drew 0.9 comments per 1,000 views.
Our samples lean toward videos that are already spreading, so read these as a high bar, not a target. How we measure.
Who belongs on your competitor list
- Four to six accounts between your size and about 10× bigger. Their audience and budget look like yours, so their mechanics transfer. Five to ten accounts in total is plenty (how to find them).
- One or two smaller accounts with a breakout. A small account’s outlier worked without a big audience behind it, which makes it the easiest kind to reproduce.
- Optional: one neighboring sub-niche account, for ideas only. If you film skincare, a hair creator with the same viewers. Borrow from their outliers, but leave them out of your benchmarks.
- Chosen by their videos, not their follower count. Search your topic in the trend finder or scan beauty TikTok trends, both ranked by views per day, and note the small creators behind fast videos.
- No celebrities or brand accounts. Their reach and ad budgets won’t tell you what would work for you.
Worked teardown: one beauty account, step by step
The same illustrative account as above. Right now fewer than two creators have 4 or more videos in our beauty samples, so this worked example stands in for the live teardown until the next refresh.
- Find the usual. The median of the twelve videos is 20,500 views. The average, 32,500, would overstate it by 59%.
- Flag the outliers. The 140K video is 7× the median of the other eleven, an outlier. The 52K video is 2.6×, a near-miss.
- Check the hit rate. Two of 12 videos cleared 2× (17%). One big hit and one near-miss is promising, not yet a pattern.
- Read the engagement. Saves rose from 13.0 to 20.7 per 1,000 views on the hit. People kept it.
- Name the mechanic. Say the hit was a concealer routine for dark circles: the finished look in the first frame, then three numbered steps in 28 seconds, voiceover, no music. The near-miss used the same structure on brows. Same mechanic twice: result first, then numbered steps.
- Write your version. Keep the mechanic, change the topic. A hair creator films the finished style first, then three numbered steps; a nail tech does the same with a chip-proof top coat. Score it against your own median after three days.
Copy the mechanic, not the topic
The topic is what a video is about. The mechanic is how it holds attention, and only the mechanic transfers.
Copying an outlier whole, with the same product, look and caption, rarely works twice: their audience has seen it, and yours can tell. Keep only what would still work on a different subject.
| What the outlier showed | The mechanic | Same mechanic, your topic |
|---|---|---|
| A concealer routine with the finished look in the first frame | Result first, then numbered steps | Finished curls first, then the three steps to get them |
| “Three brow mistakes that make you look tired” | Name a mistake the viewer is probably making, then fix it | “Three reasons your manicure chips by day two” |
| A get-ready-with-me told as a bad-date story | The routine is background; the story holds people | A night skincare routine while you tell a client story |
| A drugstore blush against a luxury one, half a face each | A side-by-side test with a verdict at the end | Two sunscreens, one on each arm, after a day outdoors |
Before you film, check three things. Would it work with a different product? If not, you are copying the topic. Does it sound like you? And are you changing one thing at a time, so the result tells you what worked?
What changes on beauty TikTok and beauty Reels
- Saves matter more than likes. Tutorials, routines and product lists are reference content, so log saves per 1,000 views on TikTok (the median in our beauty TikTok sample is 18.1). On Reels, use comments per 1,000 views.
- Check what carries the audio. In our beauty TikTok sample, 80% of the 145 videos use an original sound rather than a named track. If a voice carries an outlier, log its first spoken line: that line is the hook.
- Log when the finished look appears. If a competitor’s outliers show the result in the first second and their ordinary videos don’t, that is the mechanic.
- Compare within one platform. The median video gains about 18K views a day in our beauty TikTok sample and about 1.8K in our beauty Reels sample. The samples aren’t built alike, so that is no platform verdict, but it is why you never score a creator’s TikTok against their Reels median.
For what the whole niche is posting right now, ranked by views per day, open beauty TikTok trends and beauty Reels trends. Those pages show the niche; this method reads specific accounts.
A one-row-per-creator scorecard
The TikTok and Instagram competitor analysis templates log one row per video. Once a month, add a summary tab in Google Sheets or Excel with one row per creator:
creator,platform,videos_logged,median_views,best_multiple,hit_rate,saves_per_1k,shares_per_1k,er_by_views,mechanic,my_test,my_result
best_multiple: =MAX(FILTER(Videos!N:N,Videos!A:A=A2))
hit_rate: =COUNTIFS(Videos!A:A,A2,Videos!N:N,">=2")/COUNTIFS(Videos!A:A,A2,Videos!N:N,"<>")
These assume the per-video log sits on a tab named Videos with the template’s columns (creator in A, date posted in B, views in D, multiple in N) and the creator’s handle in column A of the summary tab, starting at A2. Fill each formula down. Works in Google Sheets and Excel 365.
- High hit rate, modest best multiple: a repeatable format, the best account to learn from.
- Low hit rate, one giant multiple: study the hit, but don’t plan around it.
- Hit rate rising: they found something; compare their last ten videos with the ten before.
- Your own row: your hit rate is the number this exercise is meant to move.
The free outlier finder shows any public account’s median views and every recent video at 2× or more of it, which covers median_views, best_multiple and hit_rate in one lookup.