Dune London gets personal with Conversant and sees 64% increase in ROI

Facebooktwittergoogle_plusredditpinterestlinkedinmailFacebooktwittergoogle_plusredditpinterestlinkedinmail

Dune London, one of the most influential global players in fashion footwear and accessories, has seen a 64% increase in return on investment per customer after personalising its media to real people. This has also resulted in a 33% increase in messaged conversion rate year-on-year.

The luxury fashion brand needed a media solution that upheld their strong brand values while driving new and repeat business.

“For us, just because somebody’s looked on our website, it doesn’t mean you should annoy them by only showing ads with that same product – it is not good business practice,” explains Mark Blenkinsop, Dune London’s digital marketing manager. “We’re interested in making sure there’s no danger of somebody looking at a product and then getting followed around the internet by that same product for three or four days – it’s something that annoys me and we don’t want to damage our brand.”

The brand sought a solution that could increase return on advertising spend (ROAS), paired with exact, targeted communications, and the ability to recognise potential and existing customers wherever they are. Only that way could they avoid brand damage while meeting higher revenue targets.

Armed with powerful profiles, rich data, and the ability to adapt and deliver personalised messages to consumers within milliseconds, Conversant enabled Dune London to reach millions of their customers and prospects.

“Conversant’s personalised media solution made perfect sense,” continues Blenkinsop. “People’s attention spans are now so short, you need to be in front of your customer at the right time and in the right place… it gives us an opportunity to tailor the consumer experience and make it personal and more relevant.”

Conversant targets real people, not segments or cookies, so each individual sees a message that is tailored to their specific needs and interests.

“Conversant delivered,” Blenkinsop states. “I have already recommended Conversant to other retailers. I can’t really think of many scenarios where it wouldn’t work.”

“Post purchase, we can show them a complimentary product like a bag – or build the relationship through brand messaging. If, after a period of time, we find out that this person only ever buys shoes we can use this information to tailor the ads to them. We know what to do because we have the historical data, and all of this is decided automatically, on the fly.

“We had it independently verified. In some cases, the attribution model that we used weighted Conversant as a greater positive influence in the path to purchase than Conversant’s incrementality model did.”