The Smart-Card is MyCheekyDate's own proprietary matching technology, developed internally and operating continuously since 2007. It is not a third-party platform. It is not a licensed solution. It is not a feature built on top of someone else's infrastructure.

It was built by MyCheekyDate, refined continuously across 26,000+ verified events in the last 10 years alone, and runs on real-world attraction data that no other company has access to.

This page explains exactly how it works, what the machine learning learns, and why that combination of proprietary technology, event scale, and time produces something no dating app or competing platform can replicate.

Some platforms that operate at live speed dating events use real-time optimization technology: attendees check in via smartphone, receive algorithmically assigned date rotations, and submit live feedback during the event. That feedback feeds directly into the algorithm, which adjusts who you meet next as the evening progresses.

This is a genuine and defensible approach. It solves the problem of filtering efficiently within a single evening.

The Smart-Card solves a different problem entirely, and does so by not intervening in the event at all.

At a MyCheekyDate event, every attendee meets every other attendee in their bracket. No algorithm decides who sits across from you. No rotation is adjusted in real time based on predicted compatibility. The evening unfolds naturally, without technological interference.

The machine learning comes after. Once the evening has concluded and attendees have privately submitted their selections, the Smart-Card analyzes what actually happened across the full event. Not what an algorithm predicted would happen. What genuinely occurred between real people in a real room.

A system that influences the event as it happens collects data shaped by its own intervention. It is optimizing toward its own predictions, which means the data reflects those predictions as much as it reflects genuine human behavior.

A system that observes the event without intervention collects uninfluenced behavioral data: who people actually chose, how strongly they felt that, and who chose them in return, none of it shaped by a prior algorithmic decision about who should meet whom.

Registration

At registration, MyCheekyDate collects name and email address only. No profile to optimize. No photo. No list of stated preferences used to pre-filter who an attendee will meet. The event begins with a clean slate for everyone in the room.

At the Event

When attendees arrive, they access the Smart-Card through a secure web link on their own phone. No app download required. On the Smart-Card, attendees enter a short bio: a few lines about themselves, written in the room, on the night, without the hours of refinement that a dating profile composed at home tends to accumulate.

This matters. A bio written in the room, under mild time pressure, before any conversations have begun, is meaningfully less curated than a profile optimized at leisure. It is closer to what someone would actually say if asked to describe themselves in sixty seconds before walking into a room full of strangers. That authenticity is a feature, not a limitation. It is the baseline the machine learning later cross-references against everything that happens in the room.

The Selection System

After each four-minute conversation, attendees privately rate the person they just spoke with across five tiers: a spectrum of genuine interest that captures not just whether they would like to see someone again, but how strongly they felt that connection.

The selection window stays open until midnight, removing social pressure from the decision entirely. Nobody is choosing in real time, in the room, with the other person still nearby. The result is a more honest signal than any end-of-night hand-in could produce.

Every MyCheekyDate event generates four simultaneous data streams that feed the machine learning. This is the part that separates the Smart-Card from anything else in the dating technology space: not just what it collects, but how many signals it cross-references to build a genuinely accurate picture of real-world attraction.

This is the dimension that most clearly separates the Smart-Card from every other matching technology operating in the speed dating space. Other technologies optimize one evening. The Smart-Card informs an entire ecosystem.

Because the machine learning continuously analyzes real-world attraction signals across thousands of events, it builds a behavioral profile of each attendee that is far richer than anything a single-event system or a static dating profile can produce. That profile feeds directly into every product in the MyCheekyDate ecosystem.

A swipe dataset, however large, is built from static images and short bios. A few seconds of judgment, repeated millions of times. Wide, but shallow. It has never observed two people's body language shift mid-conversation. Never seen a room's match rate move because the energy changed after 9pm. Never captured the gap between someone who says they want a good listener and what their actual selections, event after event, genuinely reveal about who they are drawn to.

26,000+ verified events across 65+ cities is a different kind of dataset. Not wider, but deeper. Each event produces four simultaneous behavioral signals that only exist because real interactions actually occurred.

The revealed preference gap is consistent and significant. Across the full dataset, there is a reliable difference between the attributes people describe in their bio and the attributes that actually predict who they select after a real conversation. This gap follows recognizable patterns the machine learning has become increasingly accurate at identifying.

Behavioral signals outperform demographic ones. How strongly someone rates another person, whether they return for a second event, how their selections shift across an evening: these signals predict genuine long-term interest more reliably than the static attributes a profile-first system would weight most heavily.

Accuracy compounds over time. Attendees who return for a second event see a 77% improvement in match rate over their first. With 26,000+ events in the dataset, that compounding has produced significant accuracy gains across the full system.