The problem with online property searches

In 1966, the legendary adman David Ogilvy set out to buy a home in France. He boarded a transatlantic liner to meet a French estate agent who had a perfect house waiting for him in Paris, but while still in mid-ocean he heard he had been gazumped.

There were presumably other houses on sale in Paris at the time, but it seems the agent did not show David any of them. Instead he suggested they board a train to Poitiers, 200 miles away, to an area David later described as ‘the South Dakota of France’. On the banks of the Vienne stood a decaying 13th-century château with around 30 bedrooms and a network of dungeons. David bought that instead. He lived there from the mid-1970s until his death in 1999. Restoring the building and gardens was one of the great joys of his life.

I have always wondered who that estate agent was. Perhaps he or she never had any intention of selling the house in Paris and this was a canny example of what salesmen call a ‘bait and switch’. After all, it is relatively easy to find buyers for townhouses in Paris, while only a tiny niche can be persuaded to buy a dilapidated château in the back of beyond.

I was reminded of this story when talking to the boss of one of London’s largest estate agents. I have long instinctively felt that online property search is a very bad way of buying a house, and he was able to explain why. An online search merely shows you what you think you want. When people buy through a human estate agent, however, they very rarely end up buying something matching the specifications they set in advance.

In decisions of this kind, the search process is highly recursive. We adapt our preferences on the basis of what we find. We may think we want a garden flat in Fulham but once shown a penthouse in Rotherhithe we revisit our assumptions. A good salesman may know you better than you know yourself.

This explains why economists discount the value of salesmanship. An axiom of economics is that we all have stable, known preferences. But there’s a problem with this assumption — it’s rubbish. Often we can’t get what we want, so we settle for wanting what we can get (the fox and the grapes). Or else new information mid-search leads us to reshape our desires completely.

Moreover, unnoticed by the economist, the human salesman performs a valuable function which a property website does not. He does not ask only ‘What do you want?’ but also: ‘What can I sell you that other people would not buy?’

This is a vital question to ask if markets are to function efficiently. As David Ogilvy’s estate agent knew, successful property matching is not just a question of finding properties people want, but also finding buyers who will buy things other people won’t. It devalues a property if it is next to a cemetery, beside a pub, next to a railway line or in a poor school district. But all these things are desirable to me, since I am an unsuperstitious, bibulous train--spotter whose children have left school. Ideally I’d like to search online for a Georgian rectory next to the Railwayman’s Arms in the catchment area for St Crackhead’s Academy (Ofsted rating: cataclysmic), but I can’t.

The problem when we buy or choose online (at Amazon, Rightmove, Ocado), or when we buy anything using a decision tree, is that we never get to see what we might have bought had we chosen differently.

This is why you should frequently make the effort to visit a bookshop, browse at a supermarket, or even talk to a French estate agent. Simply looking for what you already believe you want feels logical, but it’s also incomplete. Leave some room in your life for serendipity.