The market in 2026
Incumbents are flat. Money is moving to higher prices per payer, niches and real-life formats, not to more swiping. The dating app market is roughly USD 10 billion a year.
| Player | Period | Revenue | Paying users | What it signals |
|---|---|---|---|---|
| Match Group | Q2 2026 | USD 853M (−1%) | 13.3M (−6%) | Revenue held up by a higher price per payer |
| Tinder | Q2 2026 | USD 457M (−1%) | 8.5M (−5%) | Monthly users down 7%; moving to AI picks and events |
| Hinge | Q2 2026 | USD 204M (+22%) | 2.0M (+17%) | "Designed to be deleted" intent wins |
| Bumble | Q2 2026 | USD 210.5M (−15%) | 3.16M (−16%) | Two rounds of layoffs; dropped "women message first" |
| Grindr | Q2 2026 | USD 138.1M (+33%) | 1.4M (+16%) | A niche with a premium AI tier |
| Timeleft | 2025–26 | about EUR 18M a year | 150K diners a month | Dinners with strangers; figures not independently verified |
People lack connection, not matches. About 78% of app users report burnout, and the top cause is failing to find a real connection. People swipe about 58 times before one meaningful conversation and delete and reinstall apps about six times a year.
What is working
- Intent over volume. Hinge grows on prompts and comments rather than raw likes.
- Doing it together. Group formats lower the pressure: Tinder's Double Date, run clubs, small dinners.
- Verification as a feature. Tinder's Face Check cut exposure to bad actors by over 60%.
- Moving people offline. Dating apps are pushing events because users ask for them.
What failed, and why
| What | What happened | What Peoplarium took from it |
|---|---|---|
| Public video roulette | Omegle shut down in 2023 after a lawsuit over matching a child with a predator | Verified adults only, voice before video, live AI moderation |
| Video-first dating | Houseparty closed; Filteroff was sold; video feeds stalled | Video only as a consented step, never a feed |
| AI that pretends to be you | Apps where AI avatars chatted for users closed; users distrust AI-written messages | AI drafts, the member sends; AI never impersonates |
| Apps with no reason to return | Thursday closed its app to run events only; novelty apps lost most users within months | Plans, crews and meet-again carry continuity |
| Fake users and weak security | IRL had 95% fake users; Tea leaked ID images and messages | Verify everyone; keep results, never ID images |
| Pricing that felt like a trap | FTC's USD 14M order against Match; a USD 60.5M age-pricing settlement | One price for everyone, one-tap cancel, no paid visibility |
What shaped the design: the law
Four jurisdictions matter for a Vancouver launch with visitors from anywhere. The pattern is the same in each: a platform owes duties on content, data and how it responds to reports, and the more it controls what happens offline, the more it owes.
Canada and British Columbia
- No statutory safe harbor. Negligence duties grow with control. A party that creates a risk and invites people into it can owe a duty of care. This is the single biggest reason the platform provides tools and never runs meetups.
- Online harms bill (C-34). Introduced in June 2026; it would add a duty to act responsibly for covered services. Which services it covers is left to regulations.
- Synthetic intimate images. C-16 received royal assent in June 2026 and brings realistic synthetic intimate images under the Criminal Code.
- BC Intimate Images Protection Act. A tribunal can order removal and award damages against a platform. The same takedown pipeline serves these orders.
- BC auto-renewal rules. In force from 1 August 2026: cancel-anytime with a refund for monthly plans.
- Privacy and biometrics. The federal privacy commissioner's biometrics guidance asks for express consent. Quebec requires a biometric database to be declared 60 days before use, so verification is off in Quebec at launch.
- Mandatory reporting. Child sexual abuse material must be reported to Cybertip.ca. This belongs in the MVP, not later.
United States
- TAKE IT DOWN Act. Platform duties since 19 May 2026: remove reported intimate images, including copies, within 48 hours; the report form must be open to non-users.
- Section 230. Courts have protected dating apps against claims about users' offline harm. Protection weakens when a platform makes promises it breaks, or builds a feature that causes the harm itself. So the product never says "safe" or "background-checked".
- CSAM reporting. Reports go to NCMEC. A human views a hash match before a report is filed.
- FTC. Guarantees, hard cancellation and chargeback lockouts have already cost a market leader USD 14M.
European Union
- Digital Services Act. Notices, takedowns and a statement of reasons for every moderation action. Appeals need a human; small enterprises are exempt from the internal appeals system while they stay small.
- AI Act, Article 50. People must be told when they deal with AI, and AI images must be marked. Peoplarium generates no images of people in Phase 1.
- GDPR Article 22. People can contest decisions made only by automated means.
United Kingdom
- Online Safety Act. Illegal-content duties under Ofcom's codes, with a named accountable person.
- Creating a deepfake. Since February 2026, creating or requesting an intimate image of an adult without consent is an offense. One more reason AI never generates images of members.
Not legal advice. Laws here are described as what shaped the design, as of the date on this page. Check with counsel in your own jurisdiction.
What the analysis changed
- The concept moved from "an app" to "rooms over profiles": small plans and hubs, not a feed.
- Verification became the entry ticket, with the face optional.
- Video moved behind consent; roulette left the plan.
- Any feature where the platform controls the offline meetup became a liability to design out.