Account overview
Small Desk Studio @smalldesk.sample Fictional sample · Desk setups and everyday productivity gadgets
- Videos reviewed
- 16
- Every video in the sample account.
- Views at snapshot
- 1.68M
- Sum of cumulative views across all 16 videos (1,683,700).
- Engagement rate
- 11.2%
- (Likes + comments + shares + saves) ÷ views, pooled across 15 of 16 videos (left out: v12, a count is missing).
- Median 3-second hold
- 66%
- Share of views still watching at 3 s; median of 15 videos (v08 not captured).
- Publishing window
- Jul 8, 2026 – Oct 12, 2026
- 96 days, first to last video.
- Metrics snapshot
- Oct 15, 2026
- All counts are as of this date. Newer videos have had less time to collect views.
3-second hold by opening type
How many viewers each kind of opening kept past the first 3 seconds.
| Group | Videos with data | Median | Values |
|---|---|---|---|
| Problem call-out | 4 of 4 | 70% | v03 71%, v07 74%, v11 66%, v14 69% |
| Result first | 4 of 4 | 71% | v01 72%, v05 68%, v09 70%, v16 73% |
| List / promise | 2 of 3 | Not compared (too few videos) | v02 63%, v13 64% |
| Question | 2 of 2 | Not compared (too few videos) | v06 58%, v12 65% |
| Slow intro | 3 of 3 | 42% | v04 39%, v10 44%, v15 42% |
- List / promise: 3 videos, 2 with 3-second hold data. At least 3 are needed before this category is compared on its own.
- Question: Only 2 videos. At least 3 are needed before this category is compared on its own.
What the review found
Comparisons across all 16 videos. Medians, at least 3 videos per group.
- Pattern in this sample
Slow intros held fewer viewers past 3 seconds
3-second hold: slow intros 42% (n = 3) vs direct openings 68.5% (n = 12)
See the evidence - Pattern in this sample
Hooks with a specific number earned more views in their first week
First-7-day views: number in hook 121.5K (n = 6) vs no number 69.2K (n = 9)
See the evidence - Pattern in this sample
List videos were saved more but finished less
Save rate: list openings 5.7% (n = 3) vs other openings 1.6% (n = 12)
See the evidence
How this review works
Calculated in your browser
Totals, rates, medians, and group comparisons come from local rules over the sample records. So do the per-video flags (greeting-first openings, specific numbers, distance from the account median).
Prepared for this sample
The per-video feedback, example rewrites, and the wording of each interpretation were written ahead of time for these 16 records. No AI model or API runs while you use the demo.
Not analyzed
The records hold text and numbers only. Nothing here looks at footage, audio, or editing, and nothing connects to TikTok.