David Excell CEO of Featurespace

The traditional corner shop is getting a new lease of life in cyberspace – thanks to innovative software from Cambridge company Featurespace.

It is helping e-commerce businesses understand their individual customers and their buying habits.

The aim is to help them spot fraudulent activity on their sites – and also tailor their marketing material more precisely.

“Traditionally, when you had more small corner shops, the person who owned that shop got to know everyone who came in – they knew their personality, how many kids they had and so on,” said Featurespace chief executive David Excell.

“But in the online world you have hundreds of thousands – or even millions – of customers who are just identity numbers and you don’t know the personality of any of them. We’re trying to uncover what that personality is.”

Getting to know your customers helps build brand loyalty and means business decisions can be more effective.

“If, for example, you understand that someone is going to come into your shop and buy their newspaper every Saturday morning and that’s all they ever want, then don’t go marketing everything else to them because you’ll end up annoying them,” said Mr Excell.

“It’s getting that balance right – recognising the people that you want to go and sell lots to, and making marketing material as relevant as possible to them.

“A lot of marketing people will say they already send relevant information – but until now it’s been based on you being similar to someone else, rather than you being you.”

Featurespace began life in 2005 when Mr Excell was a PhD scholarship student at Cambridge University and became interested in the fields of artificial intelligence and machine learning.

“Every time I was riding from where I was in college to the university department I started to see the same people walking down the street or riding their bike at roughly the same point at the same time every day,” he said.

“People are very repetitive in the things they do so you can start to apply models around what people are likely to do – and predict what they are going to do next.”

Six years on, the company has moved from its original home in Mr Excell’s garden shed to premises in Regent Street and recently raised £1 million in investment. It has 14 employees – and plans to double in size over the next 12 months.

Its customers so far include gaming firms and online finance organisations. But the technology has the potential to cover a variety of e-commerce activities – as well as counter-terrorism applications.

The key to its success is that it builds individual profiles of customers – learning who they are and what they’re likely to do.

“You are effectively observing someone interacting and it’s a question of understanding why they are doing something – understanding the intent of the data you are seeing,” said Mr Excell.

This interpretation process is something that comes more naturally to people than to computers.

“If you see someone walking down a corridor, based on how they’re moving you can tell whether they’re in a good mood or a bad mood,” said Mr Excell.

“If they’ve got a spring in their step, they’re likely to be happy. But if they’re slouching then maybe they’re not feeling well.

“Perceptually that’s very easy for a person to understand – but to translate that into a computer is difficult.”

With revenues last year of just under £1 million, Featurespace is obviously speaking the right language. And the company is keen to allay any fears of Big Brother tactics.

“We are only using information that a business can capture about you legally anyway,” said Mr Excell. “It’s data that’s already there – we’re just intelligently deciphering it.”