John’s Dating Funnel — seven apps, one gold thread story mode ▸

The portfolio view. Five funnels, five strategies. Bumble was volume (4,006 swipes, $139.99). Sitch was an AI matchmaker (160 candidates, $159.99). Tinder was Platinum (1,662 swipes, $119.99). CMB was beans (2,616 bagels, $326.33). Hinge was money: $1,265.69, of which the export admitted $179.99 (a sub bought 24 minutes into the account) and Apple’s receipts supplied the other $1,085.70 in Superboosts and roses. Biggest tab on the board. Found in a screenshot. All five drain into the same gray pool: 3,379 people in, relationships: 0. Thirty first dates (12 CMB, 9 Hinge, the only app that asked me afterward if we’d met; I said yes seven times). Ten second dates. Three thirds, all Hinge. One relationship offer, declined: Hinge’s silver thread crawls further right than anything else here and still ends in the pool. Spend so far: $2,011.99, plus Tinder’s boosts, price still unknown. $67.07 per first date league-wide: CMB $27, Bumble $35, Sitch $40, Tinder $120, Hinge $141. Cost per relationship: a clean division by zero. The gold thread, Instagram → Chelsea: one DM to someone I met at Baylor twenty years ago. It bypasses the entire industry. (Drawn at 32× scale. At true width it vanishes. Fitting, but unreadable.) Two ribbons still buffering, both already billing: The League ($94.98) and Raya ($217.92 in Skip-the-Waits, receipts starting Jan 21, before the blitz). And two apps that will never testify: Feeld and Baroo. Accounts deleted. $234.90 in receipts is all that’s left of them.

The whole Bumble funnel, no placeholders. Nine months. 4,006 swipes, 402 messages, 23 asks. Four first dates. One second date. Then never again. Liked-to-relationship conversion: 0 for 960. Premium ran $139.99 for three months, so $35.00 per first date: best unit economics on the board until CMB did it for $27.19. (Hinge held the belt for one evening at a sub-only $20.00. Then its Apple receipts surfaced and repriced it to $140.63. Dead last.) None of it changed the outcome. The gold thread, Instagram → Chelsea, is over on the All apps board, waiting for the other six apps to pour their ribbons into the same pool.

The AI matchmaker that texts like a friend. The engine ran 36 times, generated 160 candidates, and pitched me 64. I exchanged 488 messages with the robot (101 of them mine) and about 20 with actual women. That ratio is the whole app. Nine matches (“connections,” in the billing language): five on free promo credits, four at $20 apiece. They produced four first dates, and Sitch never found out; its own date-followup table is empty. Same four dates as Bumble, on 64 pitches instead of 4,006 swipes. $40.00 per first date, third behind CMB ($27.19) and Bumble ($35). Zero second dates. $80 of unused credits stranded in the account, and two of the nine women have since deleted theirs. The gold thread, Instagram → Chelsea, remains undefeated. It has its own tab now.

The app that shreds the receipts. Fresh account on Feb 5, the day of the great app blitz. 1,334 swipes in February. 306 in March. Four in April. Silence. Platinum ran $119.99 for three months, plus four boosts, two superlike packs, and a superboost whose receipts are still at large (they never showed in the Apple history that blew up Hinge). So $119.99 per first date is a floor, and the number can only go up. It was the efficiency loser of the board until Hinge’s audit came in at $140.63. The day counters say 60 messages out, 54 back: real conversations happened. But Tinder deletes the thread when a match disappears, and 27 of my 30 are gone, every word with them. What survives: three matches, my three openers, zero replies, and one superlike note asking if we’d matched back in 2016 in Austin. The one first date happened in a thread Tinder deleted. The app destroyed the evidence of its only success. No second date. The gold thread, Instagram → Chelsea, remains undefeated. It has its own tab.

The app that actually worked, right up to the same cliff. Third account of blitz day. 2,616 bagels served, 871 liked, 77 matches, and in 53 of them she liked me first. The conversations were real for once: 363 messages out, 369 back. Parity. Nowhere else on the board. Plus 347 messages from CMB itself, mostly narrating the timer it puts on every chat: all 77 expired, and reopening one costs beans. I paid twice. It came to 12 first dates (more than Bumble, Sitch, and Tinder combined) and 4 second dates; only Hinge went deeper. Also: four women wrote first and never heard back from me. The ghosting went both ways. I remembered spending $79.98. The receipts say $326.33: a Premium month on day one, Platinum by day two, and seven bean packs ($189.93) funding 119 paid likes, two chat resurrections, and 2,457 beans I will never spend. In-app currency is spending you don’t remember doing. Even so: $27.19 per first date, best on the board. (Hinge borrowed the belt for one evening at a sub-only $20.00. Apple took it back.) And the funnel still dies at relationships: 0. The gold thread, Instagram → Chelsea, remains undefeated. It has its own tab.

The deepest run the industry got. The account predates the blitz: opened December 21, subscription 24 minutes later. $179.99 for six months, billed as a “Resubscribe,” because this wasn’t even my first Hinge era. The grind: 1,239 profiles, 748 yeses (73 with comments), 491 incoming likes X’d, 77 matches, 49 of them she liked first. I sent 302 messages, 22 of them voice notes. Only my half survives: Hinge exports the outbox and shrugs at the rest. 42 real conversations. 18 asks. Then the longest tail on the board: 9 first dates, 5 second dates, 3 third dates, one relationship offer. Declined. The only funnel that reached the final boss, and the answer was no. Hinge also took attendance: its “We met” survey logged 7 of the 9 dates, then followed up with “was she your type?” 5 yes, 2 no. One of the nos was the 61-message thread. I basically stopped swiping in February (2 likes ever after), yet 34 more matches rolled in March through May, which felt flattering until Apple explained it: $1,085.70 in consumables the export never mentioned. 21 Superboosts, one every two and a half days from Jan 20 to Mar 13. 2 Boosts. 172 roses ($415.92; fifty of them bought March 24, three weeks after my last like, straight into the stranded-currency graveyard with Sitch’s credits and CMB’s beans). The zombie spring wasn’t organic. The profile was coasting on a paid tailwind. True total: $1,265.69. I remembered $180. The receipts remembered seven times that; CMB’s 4× audit was amateur hour. $140.63 per first date, worst on the board. Tinder exhales at $119.99. The gold thread, Instagram → Chelsea, remains undefeated. It has its own tab.

The control group. Five apps in: 3,379 candidates, 317 matches, thirty first dates, $2,011.99. The industry’s high-water mark: one relationship offer, declined. Relationships: 0. Instagram: one DM to someone I met at Baylor twenty years ago. $0.00. Relationships: 1. The entire dating industry, outperformed by a friendship and a follow button. Two more ribbons are coming (The League, Raya). The score is not expected to change.