How data is shining a light on global property markets
The property market, like that of gold and oil, is a rather murky world.
The prices you'll see on most websites are asking prices. The value of a done deal - the real price - can take land registries weeks to process, by which time a fast-paced market will have moved on.
So those on the inside doing the deals, such as estate agents and developers, have a distinct advantage.
Could technology help blast open this closed market?
Teun van den Dries, chief executive of Dutch software company GeoPhy, believes his data analytics software program could do just that, starting with commercial property, a global market worth about €22.5tn (£15.7tn), according to the European Public Real Estate Association.
His program crunches lots of different data sets - public transport, roads, congestion, location, demographics, local economy, building quality and so on - to calculate an estimated value for a property.
And he has data for 41 countries, from Singapore to Spain, Brazil to Belgium.
"If you look at the current property market, almost all transactions are handled by estate agents that will describe property as being well situated, with great accessibility and beautiful views," he says.
"And that could all be true, but it doesn't mean anything and it doesn't allow you to compare."
Location accounts for 70%-75% of the weighting in the algorithm - a mathematical set of rules - and his pricing is accurate within about 5%, he says.
Estate agents are known for their creative euphemisms when it comes to property descriptions, but data could help cut through the sales speak to arrive at a more realistic assessment, he believes.
But, he notes, "a valuation is never right until someone pays. So, it's the same price point a surveyor will put their signature on."
The only difference is that it's derived from data and a set of comparable rules, he says.
However, there are some valuations it can't help us to understand - parts of London, such as St James's Park or Mayfair, home of the £90m mansion, simply defy data analysis.
At present, his customers are pension funds and other large institutions that own property portfolios. They want quick access to property valuations, as well as other data, such as the energy efficiency of their buildings.
But he hopes this type of analysis could also help make the residential property and rentals markets more transparent, too.
So when your landlord says prices are rising in your area and hikes up your rent, you'll be able to see if that's really the case, says Mr van den Dries.
What do clients think about this data-driven approach?
"It's an acquired taste in a way," says Hans Op 't Veld, head of listed real estate at PGGM, a Dutch asset manager which looks after about €155bn (£110bn) of investors' money.
"We are not that used yet to using data that massively improves understanding and transparency in the market," he says. "We are used to operating in a rather opaque market and that is changing rapidly."
He adds: "For people who are uncomfortable with that, I'd say you have to shape up... this is a trend that is unstoppable, really."
What does this all mean for estate agents?
"The biggest challenge facing the property industry generally, is that whoever controls the most amount of data, in theory, can be the most powerful, and we've always prided ourselves on having the most data," says Michael Davis of property services company JLL, which as well as being an agent also provides advice and construction services.
His firm understands the attraction of data and it is also developing its own software to help fund managers benchmark the buildings they own.
But Mr Davis believes that so long as people are parting with large amounts of money - sometimes hundreds of millions of pounds for large commercial buildings in the world's metropolises - a human will be needed to provide advice, assurances and a dash of local intelligence.
A computer can't do that, he says.
For example, a good agent will know whether a particular tenant is more or less creditworthy than recent data suggests. Thus, while the shop window part of the industry may become less profitable, areas such as advice, planning and sustainability will still be needed, he believes.
For him, access to data would mean more competition from small firms but would lead to a more efficient marketplace.
"All our property listings would benefit from as many people as possible looking at them," he says.
'Unwilling to change'
But not everyone is so sure about the benefits of data analytics in the commercial property market.
The diversity of commercial buildings makes them hard to compare, says Robbie Duncan, a property analyst at Numis Securities. And property investors will buy and sell buildings for differing reasons.
For example, a seller may offload a building to make a loss to offset against tax and as such will sell at a lower "rational" price, he says.
And shifts in economies thousands of miles away - China or in the Middle East, perhaps - could suddenly empty money out of a given market, without the data giving any warning.
While many large publicly owned property owners have talked about using data, many "just don't really know where to start and are only at the start of the journey," he says. "Commercial property is the last imperfect market."
"Homes may be better, as they are more homogenous and could be more comparable," he adds.
Mr van den Dries admits that there is some resistance to this new data-driven approach - a number of property owners have expressed displeasure at having their buildings benchmarked, he says.
But he, and others, remain convinced that better analysis of more data is key to a more efficient - and less mysterious - property market.