Both online and offline data sets have been combined to replace basic analytics tools, allowing the rapidly-growing service provider to improve end-to-end customer journeys.

Launched in 2005, Zopa allows individuals to bypass the banks to get a better rate by matching savers with borrowers via the internet. Zopa has experienced fast growth over the last few years, presently boasting 46,000 active savers and 71,000 borrowers. When customer numbers get this big, more basic business insight programmes - Google Analytics and similar tools - must be left behind.

Using the Clojure programming language, ThoughtWorks was able to quickly get to the core of Zopa's data sets and validate or reject hunches and suspicions about a wide variety of issues from customer interaction and important lending times to how customers can be offered the best rates. Myths debunked and insights gained, Zopa is now working on a range of new products and services to deliver an even better customer experience.

Dave Elliman, Market Technical Principal said: "When businesses grow quickly, they can easily be overwhelmed by the vast amount of data that becomes available overnight. It was time for Zopa to leave the Google Analytics tool behind and immerse itself in a sea of data from a wide variety of sources. Now it can push on and start to reach its true potential."

Matt McGuire, CTO, Zopa added: "Almost £500 million has been lent via Zopa and we are now brilliantly placed to lend the next half a billion. With new products and services in the pipeline, we will maintain and build upon our position as the UK's leading peer-to-peer lending service."