Happn's AI Curation: A Bold Bet Against Infinite Swipes

Dating platforms are reaching for artificial intelligence as an antidote to user fatigue, but the treatments on offer differ sharply. Happn, the Paris-based proximity-based dating app, is betting that less is more: its CEO Karima Ben Abdelmalek has outlined plans to combat burnout through AI-curated daily matches capped at 10 profiles, a sharp departure from the infinite scroll that defines most competitors. The shift matters because retention, not acquisition, has become the defining challenge for dating operators.

Users are walking away not because they can't find matches, but because the sheer volume of choice has become paralysing. Abdelmalek's pitch is that algorithmic curation can cut through the noise—though whether a smaller haystack actually helps you find the needle is the £39.5M question.

The Industry Imperative

Happn's 10-profile cap is a direct rebuke to the more-is-more philosophy that's driven dating product design for a decade.

Whether that's courageous product thinking or a feature constraint dressed up as innovation depends entirely on match quality—and there's no independent data yet to judge that. What's clear is that the industry has collectively decided AI is the answer to fatigue, even as operators pursue radically different implementations. We're watching a live experiment in whether users want algorithms to choose for them or simply help them choose better.

Create a free account

Unlock unlimited access and get the weekly briefing delivered to your inbox.

Register free

No spam. No password. We'll send a one-time link to confirm your email.

Curation as Retention Strategy

Happn's approach centres on algorithmically selecting 10 profiles daily based on location data and user behaviour patterns, according to Abdelmalek's recent statements to industry press. The cap is deliberate. By limiting volume, the company is attempting to increase perceived value per match and reduce the cognitive load that research has repeatedly linked to dating app abandonment.

The model inverts the dominant logic. Tinder built an empire on infinite swipes. Hinge grew by promising better matches but still fundamentally offers continuous browsing. Happn is arguing that the scroll itself—not just match quality—is the problem.

This isn't entirely without precedent. Coffee Meets Bagel attempted something similar years ago with daily curated matches, though it never achieved scale comparable to the major platforms. The difference this time is that AI models trained on massive behavioural datasets can theoretically make better selections than earlier algorithmic efforts. Theoretically.

Beyond match curation, Abdelmalek outlined safety applications: using AI to detect fake profiles, flag inappropriate content, and identify harassment patterns. These are table stakes, not differentiators. Every major platform from Match Group (MTCH) subsidiaries to Bumble (BMBL) is deploying similar tech, driven both by genuine trust concerns and regulatory pressure from frameworks like the UK Online Safety Act (OSA). The real test isn't whether Happn uses AI for safety—it's whether the implementation actually works better than competitors'.

The Industry Splits on AI Deployment

The curation strategy puts Happn at one end of a spectrum that's rapidly defining competitive positioning. Grindr (GRND) is building what it calls 'Grindr Wingman', a conversational AI that helps users craft messages and navigate interactions—essentially an always-available dating coach. That's assistance, not constraint.

Match Group has been more surgical, adding AI features across its portfolio but stopping short of limiting user choice. Tinder introduced AI-powered photo selection tools. Hinge added conversation prompts. These are optimisation plays, not philosophical pivots.

The split reflects genuine uncertainty about what users actually want: fewer options presented better, or infinite options with better tools to navigate them.

The answer likely varies by user cohort, relationship intent, and how burned out someone already is when they arrive at your app.

What's notable is that nearly every operator is framing AI deployment through the lens of solving user problems—fatigue, safety, match quality—rather than the operational efficiencies AI might offer. That's partly genuine product thinking and partly defensive positioning as the industry faces sustained criticism over engagement tactics that prioritise retention over outcomes.

The Monetisation Question Nobody's Answering

Abdelmalek also floated a vision of integrating with local businesses—restaurants, cultural venues—to turn matches into real-world meetups. The framing is about moving people offline faster, which aligns with stated user desires. The subtext is monetisation through partnerships and affiliate revenue.

This isn't new territory. Bumble tried variations on this with Bumble BFF city guides and venue partnerships. The challenge has always been execution: users want suggestions that feel native to the dating experience, not bolted-on advertising. Whether AI can make those recommendations feel genuinely useful rather than extractive remains unproven.

The hospitality angle also reveals something about happn's strategic position. The app reports 130 million users globally, according to company figures—material scale, but nowhere near Tinder's estimated 75 million monthly actives or Bumble's 42 million. For a platform that hasn't cracked the top tier by user volume, alternative revenue streams beyond subscriptions and à la carte features become more critical.

What Actually Changes

The broader industry thread here is that AI adoption is accelerating across dating platforms not because operators have clear evidence it works, but because they're running out of other levers to pull. Organic growth has stalled for the major platforms. Match Group reported Tinder revenue down 5% year-over-year in Q4 2024. Bumble's paying user base barely grew in the same period. The product playbook that worked from 2015-2020 has stopped working.

Curation might help. It might also just shift user frustration from "too many bad matches" to "the algorithm doesn't understand me." Early data will be critical, and operators should be watching whether happn can demonstrate improved session frequency or reduced churn among users seeing curated feeds versus traditional browse experiences.

For now, the industry is hedging: investing in AI capabilities whilst maintaining existing interaction models, waiting to see which approaches gain traction before committing fully. That's rational given the capital constraints most operators face post-valuation correction, but it also means the user experience remains caught between paradigms—neither fully algorithmic nor fully user-directed.

The question that will define the next product cycle is whether dating platforms trust their AI enough to actually constrain user choice, or whether curation remains optional flavour text on top of the same infinite scroll. Happn's 10-profile cap suggests at least one operator is willing to find out.

D