Video Summary - I Made 5 Fake Dating Profiles to See Which Aesthetic Wins

What the video does (core experiment)

The creator runs a controlled dating-app experiment on Hinge by creating five fake dating profiles of the same person (described as “four of me,” later becoming three successfully launched profiles plus one cancelled/paused in the end stages).

Goal: test whether aesthetic differences (photos + prompts) change:

How many likes/comments a profile gets
What types of men are attracted (politics, lifestyle, tone)
Red-flag rates inferred from messages/comments

Method: launch the profiles at the same time to collect comparative engagement data over ~24 hours, then build a presentation from tracked results.

Aesthetic “characters” created

1. Neutral / Control (dropped early)

Whitney Fugal (neutral daily-life aesthetic)
Removed after Hinge biometric face verification fails/scares the system

2. Girly city girl (Melanie Shik)

Urban, “effeminate” city vibe: gym, matcha, cafes, performative lunches
Concern: could appear high maintenance / intimidating, potentially scaring away certain users

3. Tomboy / boyish sporty girl (B Toppins / “Bee”)

Baggy sporty sneakers, minimal makeup; candid, less “posed”
Goal: attract more “equal/peer” style interactions; avoid “stunting” and performative posing
Notable risk: some profile behavior/themes (e.g., weapons/“drink/pool/FPS” references) may negatively affect match quality for some men

4. Trad wife / cottagecore traditional (Tegan McKenzie / “Soft”)

“Soft lifestyle” trad vibe: orchard strawberries/chickens, submissive imagery
Expected: long-term serious partners
Results show lower comment engagement

5. Vamp / femme fatal vampire-red aesthetic (Jade Somerset)

Dark, seductive, “witchy”/goth-adjacent tone
Photos include castle/occult/architecture/cemetery-style settings
Thesis: test reaction to confident intelligence
Results suggest fetishization without equal respectful engagement

Hinge constraints + anti-fraud/verification workarounds (technology details)

Anti-catfish friction

Hinge makes it hard to run multiple accounts; women’s profiles get locked out more than men’s
The setup required multiple phones + phone numbers + accounts
Because the experiment needed full visibility into likes, the team used Hinge Plus subscriptions, apparently requiring multiple credit cards (free accounts limit like viewing)

Mandatory biometric face check rollout

Hinge introduced biometric face verification (“face check”) to prevent fraud/duplicate accounts
The creator tried to “fool” it using:
- different lighting
- facial expressions
- facial positioning
- (e.g., “Play-Doh on nose,” puffing cheeks/lips)

Troubleshooting included:

Swapping photos’ metadata/EXIF behavior
- photos taken on one camera, edited on one computer could trigger similarity detection
Exporting/handling metadata so profiles appear from different upload conditions/devices
- intended to reduce suspicion before human review
Using different phone service providers/towers
- two phones on different networks, with verification before/after uploading

Verification outcomes

Three profiles verified successfully
One profile (“Melanie”) failed verification initially because her profile wasn’t fully finished
Verification timing/conditions affected the experiment’s progression
They contacted Hinge support only as an emergency workaround
Ultimately they had to delete/pause/recreate in response to verification issues

Photos + prompts as “feature vectors” (how the profiles were designed)

Each fake profile included:

A full set of profile photos (dozens captured across “basic Toronto places,” edited for consistency)
3 prompts per profile, written to match the character identity

Prompt categories (personality proxies) included:

music taste (e.g., “favorite punk or metal album in high school”)
interests (e.g., video games vs indie music)
lifestyle categories tied to each aesthetic

Data collection + metrics analyzed

Engagement was recorded across:

likes
comments
conversation conversion rate (comments interpreted as deeper interest)
premium engagement: roses on Hinge (premium currency)
inferred user attributes from profile text/comments: age, height, location, job, politics, religion, etc.

“Red flags” tagging

Red flags were tagged from message/profile content, including:

contradictions across profile vs comment behavior (e.g., claiming one political stance then commenting another)
narcissistic/vanity selfie patterns (e.g., mirror flexing)
ideological/manipulative content (examples described for the trad/wife and vamp archetypes)

Key results: which aesthetic “won” and why

Rankings (based on combined engagement + analysis; some profiles were paused/impacted by verification timing):

1st: Tomboy (“Bee” / B Toppins) — most successful overall
2nd: Vamp (“Jade Somerset”) — extremely high comments, strong lust/interest patterns
3rd: Girly city girl (“Melanie”) — strong engagement (likes/roses) but fewer total likes than earlier rivals
4th: Trad wife (“Tegan / Soft”) — lowest comment conversion despite many likes
- described as more “looked at but not engaged”

Notable interpretations of audience behavior

Bee (tomboy)

Attracted youngest likes
More “goofy/sweet/quirky musician-type” men
Lowest red-flag rate: reported at 15%
More “relatable/equal” banter due to shared interests (gaming/music/outdoors)
Tradeoff: conversations could feel less romantic/formal, sometimes casual (“joking/1v1” energy)

Jade (vamp)

High comments and attention from well-established/wealthier men
Zero roses reported, implying a gap between attraction and premium investment
Many men appeared to fetishize power/seduction rather than engage respectfully

Melanie (girly)

Strong comment conversion; received roses (including multiple roses)
Red flags lower than some profiles (stated at 18%)
Still attracted “pedestalizer/helpless/demand-skincare” dynamics (men begging or objectifying under a “high maintenance / caretaker” framing)

Tegan (trad wife)

Despite long-term suitability “on paper,” had low comment conversion (~20%)
Higher proportion of ideologically intense or “moderate-liar” behavior
- men described themselves as moderate but behaved conservatively in comments
Some matches leaned toward objectifying/submissive-provider fantasy rather than respectful partner framing

Timing effects and experiment limitations

Profiles went live near a specific window and went offline about a day later; timing may have affected results
Hinge “Boost” was used:
- “Boost” paid feature ($12.99) to place a profile in front of 11x more people for 1 hour
- Melanie was boosted later to “skew”/adjust late-stage imbalance, affecting comparison purity
Major real-world confound:
- Melanie couldn’t verify at first and required fixes while others were verified
Safety/contagion issue:
- they paused/offlined early when a friend found one profile “too close to home”

Final takeaway (what the video claims)

Aesthetics meaningfully change who swipes/matches and the kind of conversational intent that shows up
- e.g., peer banter for tomboy
- fetishized forwardness for vamp
- pedestal/begging dynamics for girly
But the profiles aren’t “just style”—the combination of photos + prompts acts like a targeting system that influences how people interpret the profile.

Main speakers / sources (end)

Primary speaker/creator: the narrator who films and creates the profiles (credited as “I” throughout)
Secondary recurring collaborator: Terry
- handles parts of filming/photos and some experiment operations (prompts, logistics, verification steps)
Hinge system/support: Hinge’s anti-fraud/biometric face verification and responses from Hinge Support (discussed, not directly quoted)

I Made 5 Fake Dating Profiles to See Which Aesthetic Wins

Key takeaways