TransferWise has built its reputation by bashing traditional banks for overcharging customers on currency transfers, branding established lenders as 'rip-off artists' in a series of brazen anti-bank advertising stunts over the past two years.

While competitors have huffed and puffed at the claims made by TransferWise - in particular its failure to mention how much it relies on incumbent bank processing and services to run its business - the company has grown to achieve unicorn status, handling more than £3 billion in foreign currency transfers on behalf of users.

TransferWise may have the most elegant and easy-to-use UX in the market, but claims made at the point-of-sale regarding consumer savings when compared to banks have been slammed by the advertising standards watchdog.

TransferWise bases its claims on mystery shopping exercises to compare competitors' costs of transferring foreign currency with the cost of TransferWise's service. The results of those mystery shopping exercises form the basis of the firm's claims "You're saving £xx" and "you save up to 90%".

In its ruling, the ASA says that TransferWise manipulated the figures to its advantage, and failed to provide consumers with direct comparative data on which they could make a judgement.

"The ad must not appear again in its current form," the watchdog concluded. "We told TransferWise Ltd not to make comparative savings claims against banks unless they held adequate substantiation for them and to ensure in future that any such claims were verifiable."

In a statement, TransferWise director of communiations Jo White says: "We made a statement on our website that customers would save 90% compared to their bank. This is based on regular, independent market research which consistently shows that the average TransferWise customer in the UK saves up to 90% or 10x what they would pay through their bank.

"The issues raised in the adjudication were that we weren’t thorough enough in communicating on our website the workings of how we arrived at this calculation and also that we used an average figure rather than provide the figure specific for each route. We have created a page on the site that explains the workings behind the calculation."