Chemistry cannot be filtered

Ask someone what they are looking for and they will give you a list. A range of ages, a certain level of ambition, a city, a view on children. The list is sincere. It is also, on the evidence, a poor guide to who they will actually fall for.

This is one of the more uncomfortable findings in the choice literature. When people are given a large set of options, the partners they select drift further from the preferences they stated at the outset, not closer. More filtering, applied to more people, produced choices that matched their own criteria less well.

Anyone who has been set up knows the feeling from the other direction. The person who looked wrong on paper turned out to be the only one you wanted to talk to. The one who matched every line of the list bored you before the starters arrived.

Criteria are not useless. They are a filter for eligibility, and eligibility matters. Whether someone wants children, whether they are actually available, whether they can build a life in the same place as you: these are not preferences, they are facts, and getting them wrong wastes years.

What criteria cannot do is predict attraction. They do not capture pace, or humour, or how someone treats a waiter, or whether they get more interesting or less as the evening goes on. They do not know that you are drawn to people who challenge you and bored by people who agree with you, which is a pattern you may not have noticed yourself.

The work of matchmaking sits in that gap. A matchmaker's value is not in filtering more precisely. It is in having spent enough time with you to see the things you did not put on the list, including the ones that contradict it. Most people, asked carefully about their last three relationships, reveal a pattern they have never articulated.

That is why the intake conversation is long, and why it is about history rather than preference. What you say you want narrows the field. What has actually happened to you tells us where to look.