Do 01
Bring Meng back next episode — explicit absence noted with concern.
EvidenceThree top comments ask: #12 dugongzzz (5 likes), #18 user-yb6tk1ru6x (3 likes), #27 chriskrone2557 (2 likes).
Watch forComments-mentioning-Meng drops to <1 next episode; comment count on Meng's return-episode > this video's 123.
Do 02
Make the 'CAN vs WANT' framing the title/hook of a follow-up — it's the moment the audience consistently quotes.
EvidenceHosts state it at 15:43-16:02 ('Vers actually means you WANT to do both'); commenter #14 surfaces it independently; commenter #7 (9 likes) calls Don's 'anytime' framing 'the crux'.
Watch forFollow-up video hits ≥18k views (130% of this one) within 14 days.
Do 03
Run a poll/community-post asking 'should profiles let you rank preferences?'
EvidenceCommenter #4 (15 likes) literally proposes the feature; matches the host complaint at 6:45 ('there should be more text for that on the apps').
Watch forCommunity-post engagement ≥10% of subscriber count.
Do 04
Cut a Short from 15:43-16:02 with text overlay 'CAN vs WANT'.
EvidenceHighest-density comment quotation in the long-form thread; clearest standalone insight.
Watch forShort ≥50k views in 14 days; ≥3% click-through to long-form.
Do 05
Do a 'gay dating outside Japan' comparison episode — multiple international commenters volunteer their context.
EvidenceComments from Germany (#16 Love_TheArtist), Romania (#69 c_cma1971), US (#31 USCanthony, 200+ words), Canada (#58 eccuk777) all compare home-country norms.
Watch forGeographic spread of comments widens; views break the channel's typical Japan-resident floor.
Do 06
Reply-thread the top 5 long-form comments (especially #31 USCanthony and #62 willujames) in-app.
EvidenceBoth are 200+ word substantive essays — engaging them publicly converts them into recurring power-commenters.
Watch forBoth commenters reappear on next 2 uploads.
Do 07
Greenlight the 'Boyfriend Season 2 review' episode.
EvidenceCommenter #2 simonpatience9540 (49 likes) — second-most-liked comment is a direct format request.
Watch forEpisode posts within 14 days; first-24h CTR ≥7%.
Do 08
Address the 'sides' confusion in a dedicated explainer — there's clear unmet demand.
EvidenceHosts themselves end the video confused about 'vers side' (17:55-18:25); commenter #17 thetaClysm 'thanks for introducing me to side, had to look it up'; #47, #46 give competing definitions.
Watch forSides explainer video ≥1.5× channel median first-week views.
Do 09
Show the on-screen Japanese app UI (9 Monsters profile fields) while interviewing — currently a missed visual.
EvidenceAudience-international (#16, #69, #58) doesn't know the local app context; on-screen UI would close the gap and add b-roll value.
Watch forAudience-retention curve flattens through the 3:00-5:00 interview window.
Do 10
Lean into Don as recurring on-camera correspondent rather than guest.
Evidence#10 dinodon2142 (6 likes) 'he's secretly the 3rd crew member'; #22 Topzyturvey on his teeth; #54 jasonbroccoli (interviewee) thanks him by name — parasocial momentum is already there.
Watch forDon-fronted segments retain ≥5pp better than baseline interview retention.
Do 11
A/B test thumbnail with the literal text 'Is vers a scam?' (commenter-coined phrasing).
Evidence#15 PokhrajRoy. (4 likes) wrote it spontaneously — it's the title audience already wrote for you.
Watch forCTR ≥6% within 72h of swap.
Do 12
Add a pinned comment with a glossary of terms (top/bottom/vers/side/vers-top/vers-bottom).
EvidenceHosts themselves end the video unable to define 'vers side' (18:25); #17 had to look up 'side'; #46/#47 give incompatible definitions.
Watch forReduces repeat 'what does X mean' comments on follow-up videos by ≥50%.
Do 13
Pitch Grindr or 9 Monsters for a sponsored explainer or co-branded survey episode.
EvidenceApps are named ≥7 times across video + comments; the entire video critiques their UX.
Watch forPitch sent within 14 days; response or referral at minimum.
Do 14
Run a viewer survey ('what does vers mean to you?') and dedicate the next episode to results.
EvidenceVideo literally ends with the hosts wanting more data; commenters are already self-surveying in the threads.
Watch forSurvey ≥500 responses; result-episode beats this one's view count.
Do 15
Stop shooting in audible cold — it visibly degrades performance.
EvidenceAndrew flags it twice (0:22, 12:05); commenters #37 and #44 reference the cold.
Watch forIndoor episode retention curve improves vs outdoor cold-weather episodes.
Do 16
Use timestamped chapters — none currently exist and the video has clear segment breaks.
EvidenceChapters: 'none' in metadata; the video has natural acts (intro / interview 1 / interview 2 / host-debrief) at 1:14, 2:53, 14:39.
Watch forAverage view duration ≥+10%; chapter-click data populates.
Do 17
Reach out to commenter #20 (B.WeirdGaming 'I'd love to edit for you guys') — recurring 'where is the editor' comments suggest production bandwidth gap.
EvidenceComment #20; channel's posting cadence inconsistency implied by 'review Boyfriend Season 2' backlog ask.
Watch forFaster posting cadence within 30 days.
Do 18
Open every episode with a one-line viewer-question call-to-action ('What does X mean where YOU live?').
EvidenceGeographic-comparison comments (#16 Germany, #58 Canada, #31 US, #69 Romania) all volunteered home-country context unprompted.
Watch forComment-geo-diversity ≥+25%; surfaces international audience for ad-sales pitches.
Do 19
Create a short follow-up addressing the 'becoming vers/bottom after 30' theme.
Evidence0:06 host claim; 4:57 interviewee echo; #39 karencrist4355 comment.
Watch forFollow-up holds ≥80% retention to the 30+ aging beat.
Do 20
Reply to commenter #5 月風-m4h (Taiwan mountains) publicly — earned 13 likes as a 'wholesome outsider' moment.
EvidenceComment #5 (13 likes) is one of the top-engagement comments and represents the long-tail international rural audience.
Watch forSubscriber gains from Taiwan IP in next 14 days.