Empirical data from 24,373 TikTok videos posted in 2026 across 2,769 creator accounts.
By Creafico Research • Updated October 2026 • 6 Downloadable CSV Datasets Included
Executive Summary & Key Metrics
To cut through algorithm rumors and creator advice myths, Creafico extracted first-party performance metrics across 24,373 clean, publicly accessible TikTok videos from 2,769 creators ($n = 21,771$ videos with share velocity data across 2,477 creators) published in 2026.
Core Benchmark Metrics
- Total Videos Analyzed: 24,373 videos
- Total Creators: 2,769 accounts
- Videos with Share Velocity: 21,771 videos
- Open Data Assets: 6 downloadable CSV tables
1. Follower Tier Baselines & View Distributions
To determine what constitutes normal algorithmic distribution versus a genuine penalty, we segmented all 24,373 videos by creator follower size. This establishes empirical view expectations across every phase of account maturity.
Baseline Table: View Distribution by Follower Count
| Follower Tier | Sample Size | Median Views | % < 200 Views | % < 500 Views | % < 1,000 Views |
|---|---|---|---|---|---|
| < 1k Followers | n = 4,430 vids / 608 creators | 327.0 | 35.1% | 61.0% | 76.7% |
| 1k–10k Followers | n = 6,644 vids / 766 creators | 817.5 | 19.2% | 38.8% | 54.0% |
| 10k–50k Followers | n = 4,339 vids / 470 creators | 4,641.0 | 6.4% | 15.1% | 24.4% |
| 50k–100k Followers | n = 1,951 vids / 202 creators | 10,179.0 | 4.1% | 11.3% | 18.0% |
| 100k+ Followers | n = 7,009 vids / 723 creators | 57,644.0 | 1.1% | 2.8% | 5.4% |
Key Takeaway: Over 60% of videos published by accounts under 1,000 followers land below 500 views (and 35.1% under 200 views). Hitting 200–500 views is not algorithmic suppression or a shadowban; it is the platform's standard cold-start testing window. Even accounts with 10k–50k followers see nearly 1 in 4 videos fail to cross 1,000 views.
2. Severe View Dips (<25% of Median) & The Bounce-Back Reality
When an established creator publishes a video that generates less than 25% of their typical views, creators commonly fear an account-level shadowban. We tracked 22,429 videos from 2,391 eligible creators (median views ≥ 100) and evaluated 1,493 consecutive post pairs immediately following a severe flop.
Severe View Dip Prevalence & Bounce-Back Metrics
| Metric | Measured Value | Editorial Insight |
|---|---|---|
| Severe Dip Rate (% of all videos) | 9.77% (2,191 / 22,429) | ~1 in 10 videos suffers a severe view dip (<25% median). |
| Creator Prevalence (% experiencing ≥1 dip) | 47.2% (1,128 / 2,391) | Nearly half of all active creators experience severe flops regularly. |
| Subsequent Post Pairs Evaluated | 1,493 pairs | Immediate post following the severe dip. |
| Median Multiple vs Flop Video | 3.53x | The next video outperforms the flop by 3.5x on median. |
| Immediately Exits Severe Dip (≥25% median) | 60.1% | 6 in 10 creators immediately recover out of the severe dip. |
| Recovers to ≥50% of Creator Median | 48.4% | Nearly half return to healthy baseline views immediately. |
| Complete Rebound to ≥100% of Median | 32.8% | 1 in 3 immediately matches or beats typical account views. |
| Turns into Viral Outlier (≥2.0x Median) | 14.3% | 1 in 7 next videos breaks out into a 2x viral hit! |
Key Takeaway: Severe flops are a universal feature of short-form recommendation algorithms, impacting 47.2% of creators. Because TikTok tests each post independently, the very next upload outperforms the flop by 3.53x on median, with 60.1% exiting the dip and 14.3% breaking out into 2x viral outliers. A single dead video does not flag an account.
3. The Diagnostic View Tiers & The "Share Gate"
To determine what triggers algorithmic scaling beyond initial audience batches, we analyzed engagement multipliers across diagnostic view brackets: <100, 100–699, 700–4,999, 5,000–29,999, 30,000–99,999, and 100,000+ views.
Diagnostic Tier Distribution & Engagement Multipliers
| View Tier Bracket | Video Count | % Dataset | Median ER | Shares / 1k Views | Comments / 1k | Likes / 1k | Median Length |
|---|---|---|---|---|---|---|---|
| < 100 Views | 1,787 | 7.3% | 3.70% | 0.00 | 0.00 | 27.8 | 31.3s |
| 100–699 Views | 5,780 | 23.7% | 6.13% | 0.00 | 2.02 | 53.3 | 39.2s |
| 700–4,999 Views | 5,733 | 23.5% | 5.15% | 1.34 | 1.74 | 46.0 | 49.4s |
| 5,000–29,999 Views | 4,158 | 17.1% | 5.16% | 1.74 | 1.01 | 46.6 | 57.5s |
| 30,000–99,999 Views | 2,150 | 8.8% | 6.03% | 2.14 | 0.79 | 54.0 | 60.0s |
| 100,000+ Views | 4,765 | 19.6% | 7.09% | 2.56 | 0.47 | 64.4 | 60.0s |
Key Takeaway: The Share Gate: Below 700 views, the median video receives literally 0.00 shares per 1,000 views. Crossing 700 views requires initial social transmission (1.34 shares/1k views), which accelerates by +91% to 2.56 shares/1k views on 100k+ videos, while likes per 1,000 views plateau. Concurrently, comments per 1,000 views collapse by 77% as content reaches broad, passive FYP audiences.
4. Posting Frequency vs. Outlier Probability
Creators frequently debate whether posting multiple times per day causes algorithmic cannibalization or increases viral surface area. We grouped 2,706 creators by active publishing cadence.
Publishing Cadence vs Outlier Hit Rates
| Publishing Cadence | Creator Count | Video Count | Median Views/Vid | Total Weekly Views | 3x Outlier Rate | 5x Outlier Rate | Exp 5x Hits / Mo |
|---|---|---|---|---|---|---|---|
| < 2x / week | 1,740 | 15,760 | 5,657 | 915 | 23.7% | 19.8% | 0.19 (1/5 mos) |
| 3–4x / week | 372 | 3,260 | 1,757 | 3,948 | 13.3% | 9.2% | 1.25 (~1.3/mo) |
| 5–7x / week (Daily) | 123 | 1,078 | 3,222 | 16,865 | 12.0% | 7.9% | 2.09 (~2.1/mo) |
| 1–2x / day | 200 | 1,831 | 1,279 | 8,918 | 11.9% | 7.5% | 3.04 (~3.0/mo) |
| 3+x / day (Hyper) | 271 | 2,381 | 348 | 17,443 | 11.9% | 6.9% | 9.21 (~9.2/mo) |
Key Takeaway: Hyper-volume posting (3+ posts/day) dilutes per-video median views down to 348 (-89% vs. daily posting), but maintains a stable ~7% independent probability of hitting a 5x outlier per video—yielding ~9 monthly viral outliers compared to ~2 for daily posters. For quality creators, daily posting (5–7x/week) represents the optimal balance of per-video reach (3,222) and weekly volume (16,865).
5. Mythbusting Consecutive Post Spacing (<24 Hours)
A persistent social media guideline instructs creators to wait 3 to 6 hours between uploads to prevent "cannibalizing" previous videos. We tested 8,425 chronological video pairs published by the same creator within a 24-hour period.
Consecutive Post Spacing & Performance
| Spacing Interval | Pairs Evaluated | Median % Creator Median | Median % Prev Video | Flop Rate (<0.5x) | Normal (0.5–1.5x) | Outlier (≥2.0x) |
|---|---|---|---|---|---|---|
| < 1 Hour | 1,822 pairs | 93.5% | 101.9% | 23.4% | 58.5% | 12.3% |
| 1–3 Hours | 1,087 pairs | 92.9% | 100.0% | 24.1% | 57.0% | 13.2% |
| 3–6 Hours | 1,019 pairs | 93.8% | 96.7% | 24.4% | 53.3% | 15.4% |
| 6–12 Hours | 1,132 pairs | 91.4% | 93.1% | 27.8% | 52.2% | 14.6% |
| 12–24 Hours (Ref) | 3,365 pairs | 90.8% | 93.3% | 24.2% | 54.7% | 15.2% |
Key Takeaway: The 3-hour waiting rule is a myth. Videos published under 1 hour after a previous upload achieve 93.5% of the creator's median views and a 101.9% ratio relative to the previous post, with no increase in flop rate (23.4% vs 24.2% for 12–24h). TikTok's algorithm distributes content to distinct viewer buckets asynchronously without penalizing rapid uploads.
6. Video Duration Sweet Spots & TikTok SEO Signals
Part A: Video Duration vs. Reach
| Duration Bracket | Video Count | Median Views | Median ER | Shares / 1k | % ≥ 10k Views | % ≥ 100k Views |
|---|---|---|---|---|---|---|
| < 15s (Micro-Hook) | 2,780 | 1,366 | 6.58% | 1.23 | 31.2% | 15.0% |
| 15–34s (Short-Form) | 6,583 | 1,710 | 6.11% | 1.28 | 32.4% | 16.2% |
| 35–59s (Mid-Length) | 3,967 | 4,123 | 5.35% | 1.36 | 39.8% | 20.7% |
| 60–119s (Long-Form) | 5,965 | 7,395 | 5.97% | 1.50 | 46.5% | 24.2% |
| 120s+ (Deep Dive) | 4,587 | 3,761 | 5.02% | 1.20 | 40.7% | 20.2% |
Part B: Caption Length & Search Intent
| Caption Bracket | Video Count | Median Views | Median ER | Shares / 1k | Comments / 1k | % ≥ 10k Views |
|---|---|---|---|---|---|---|
| < 50 chars (Minimal) | 7,551 | 2,668 | 6.96% | 1.11 | 0.83 | 37.8% |
| 50–149 chars (Standard) | 11,625 | 3,940 | 5.60% | 1.34 | 0.84 | 40.0% |
| 150–299 chars (Keywords) | 3,089 | 2,998 | 4.90% | 1.67 | 0.81 | 37.3% |
| 300+ chars (Micro-Blog) | 2,108 | 2,326 | 4.38% | 1.51 | 0.65 | 33.6% |
Key Takeaway: Videos between 60 and 119 seconds achieve 7,395 median views (5.4x higher than <15s videos) and have the highest rate of crossing 100k+ views (24.2%). Concurrently, 50–149 character captions deliver the highest raw views, while keyword-dense captions (150–299 characters) drive the highest share velocity (1.67 shares per 1,000 views).
7. Open Datasets (CSV Downloads)
All aggregated summary tables are available as open CSV downloads:
- query1_view_tiers.csv
- query2_follower_tiers.csv
- query3_posting_frequency.csv
- query4_post_spacing.csv
- query5_video_duration.csv
- query5_caption_seo.csv
8. Methodology & Privacy Safeguards
Data was aggregated from 24,373 publicly accessible TikTok videos across 2,769 creator accounts submitted to Creafico's TikTok Profile Analyzer tool. Videos were collected via structured scraper instances and deduplicated on unique video identifiers (shortcode / aweme_id). All reports and CSV assets contain cohort-level aggregates only. No individual creator handles or video URLs are published.