Predictive Matchmaking for Online Dating

One of the biggest problems faced by dating sites today is keeping visitors retained on site for as long as possible and keeping them as active as possible, providing real time recommendations of potential dates based on the implicit preferences derived from a user's activity on a site.

Gavin Potter, a world expert in this field, and CTO of Introanalytics explains: “Most people are consistent in what they do which is invaluable data to a dating site. We can show men and women more of what they like and less of what they don't like purely based on what they do and not what they say in their profile.”

One-way recommendations are great as a retention tool, but what about matchmaking? Just because I like someone doesn't mean they'll like me back. Potter continues: “This is where the beauty of this technology can genuinely match people. We can predict who will like each other even before they have clicked on each other. That is pretty important to any type of dating site, whether casual or serious.”

Martin Bysh, CEO of, is very satisfied with the results: “We are seeing an approximate increase of 15-20% in overall daily site traffic and this has a knock on effect of about 10% improvement in premium member conversions. We are now committed to trialling the real-time technology on offer to further improve results.”