2026-11-06 · 68 competitor video donors we hold with transcripts, labelled by what the first spoken line does, joined to the AdSpy and Ad Library records for run length.
Pooled across the whole category the answer is that it makes no difference: proof-first openers run a median 102 days and every other opener runs 93 days. Split by the door the brand sells on, the picture changes, and it changes in our favour on the only door that matters here.
Proof-first openers run a median 163 days across 7 donors and every other opener runs a median 82 days across 18, which means a proof-first opener lasts 2.0 times longer on this door.
| days run | opener | advertiser | engagement |
|---|---|---|---|
| 490 | compare ★ | Dr. Squatch | 879 |
| 485 | witness ★ | Dr. Squatch | 3,677 |
| 284 | claim | Atomfresh | 0 |
| 208 | witness ★ | Atomfresh | 0 |
| 192 | skit | SALT & STONE | 10,268 |
| 163 | compare ★ | Fussy | 0 |
| 162 | claim | Fussy | 0 |
| 147 | founder | Fussy | 2,745 |
Proof-first openers run a median 79 days across 7 donors and every other opener runs a median 93 days across 17, which means a proof-first opener lasts 1.2 times less long on this door.
| days run | opener | advertiser | engagement |
|---|---|---|---|
| 904 | founder | Sawyer Naturals | 0 |
| 218 | authority ★ | Kasper Kubica | 0 |
| 189 | claim | Hume Supernatural | 6,606 |
| 180 | problem | Carpe | 0 |
| 155 | problem | Carpe | 6,607 |
| 151 | authority ★ | Carpe | 4,186 |
| 108 | claim | Native | 16,096 |
| 105 | claim | MAGS Skin | 0 |
Proof-first openers run a median 71 days across 4 donors and every other opener runs a median 347 days across 10, which means a proof-first opener lasts 4.9 times less long on this door.
| days run | opener | advertiser | engagement |
|---|---|---|---|
| 622 | deal | Lume Deodorant | 13,529 |
| 448 | deal | Shannon Klingman, M.D. | 19,446 |
| 408 | problem | Lume Deodorant | 38,499 |
| 382 | problem | Shannon Klingman MD | 10,056 |
| 364 | claim | The Smell Good Company | 0 |
| 330 | skit | Manscaped | 29,508 |
| 263 | skit | Shannon Klingman, M.D. | 32,502 |
| 161 | witness ★ | Lume Deodorant | 202,417 |
A proof-first opener is a minority play in this category. The grid counts donors, and a star marks the three openers that put proof in front of the viewer before the advertiser says anything.
| opener | scent | odor | body | all |
|---|---|---|---|---|
| witness ★ | 4 | 2 | 1 | 7 |
| compare ★ | 3 | 1 | 1 | 5 |
| authority ★ | 0 | 5 | 2 | 7 |
| problem | 2 | 8 | 4 | 14 |
| claim | 8 | 7 | 1 | 16 |
| skit | 6 | 3 | 2 | 11 |
| deal | 1 | 0 | 3 | 4 |
| founder | 2 | 1 | 0 | 3 |
| street | 0 | 1 | 0 | 1 |
witness = another person reacts to the scent · compare = a price or luxury-brand comparison · authority = a credentialed person or a test · problem = the viewer's own pain · claim = a product property the advertiser states · skit = a comedy bit · deal = a sale or a drop · founder = an origin story · street = a street interview on an unrelated question.
Run length measures survival on the platform, and a brand with a large budget keeps an ad alive longer than its creative alone would. Two of the scent door's proof-first ads are Dr. Squatch, who outspends every other advertiser in this set, so budget and opener are not separated in these numbers. The labels are my reading of each opener rather than a machine's, and the body door runs the opposite way, so the effect is specific rather than general.
The Out There clip library held zero clips before today, which is what blocked every proof-first format. A YouTube search returned 165 candidates across the three classes and these are the ones pulled so far. The ones marked as having no brand on screen are the useful ones for a witness hook, because the reaction belongs to a smell rather than to a competitor's label.
Each clip plays the speaker's own words and nothing in our voiceover will say or imply that any of them uses or endorses Out There. Source, date, speaker and words are recorded in SOURCES.json.