Cookies don’t cut it anymore. They can’t be relied on when tracking mobile sales or when a customer moves between devices. Without a cross-device solution in place, sales are missed or incorrectly assigned.
The inability to accurately track purchases that occur on multiple devices effectively is holding back mobile ecommerce. Sales are occurring on mobile devices and advertisers are making money, but they don’t have an accurate understanding of how these sales are generated and the device on which they are converting. This makes it impossible to evaluate the ROI of channel and device performance and inform future marketing spend.
Cross-device tracking is essential technology for any customer-centric marketer who wants to accurately track consumer purchase behaviour, and correctly assign sales to the devices and channels on which they converted.
What is cross device tracking?
Cross-device tracking makes an anonymous connection between users and their devices. This connection creates a profile and every time one of these user profiles interacts with a website, the activity is tracked. Without a cross-device solution enabled, when a consumer uses more than one device to complete a purchase the connection is lost and the interactions are not tracked.
The ability to track across devices provides advertisers with a more accurate picture of how their sales are generated. Publishers also benefit from being rewarded for multi-device purchases.
Probabilistic versus deterministic
There are two ways to track cross-device behaviour, probabilistic and deterministic. Both methods pair a user with their devices in order to track behaviour and sales.
Probabilistic matching collects numerous data points about each device and uses complex algorithms to identify probable connections. Relying heavily on probability, this matching technique is an inexact science and due to its nature, accuracy rates can vary from as little as 60% to 90%. Although tracking that uses probabilistic matching tracks sales that would have otherwise been missed, it also generates false sales that the advertiser pays for.
Deterministic matching is near perfect. It uses first party data to match a user with their devices (login data). There are privacy issues surrounding this type of data and companies using this technique have a responsibility to protect consumers’ information.
Tradedoubler’s Cross Device Tracking technology uses deterministic matching. It anonymises the profile created by matching the user with their devices, to avoid collecting any personally identifiable information (PII). Consumers can also ‘opt out’ in a similar way to the cookie tracking opt out process.
Why are the big boys using deterministic matching?
‘Cross-device’ is one of the latest marketing industry buzzwords and everyone wants to jump on the bandwagon. The barriers to entry of creating a probabilistic cross-device tracking solution are relatively low, the tech is based on an algorithm and reach can be scaled quickly. So why aren’t the big boys, like Google and Facebook, taking this approach? Simple. They don’t need to do. Their networks are big enough to enable the significant reach required to build millions of user profiles, without the need to rely on probability.
Any company with a deterministic cross-device solution has massive reach within its own network. The companies that can’t scale a deterministic solution are opting for the next best thing, probabilistic matching.
But do I really need a cross-device tracking solution?
We are living in a multi-device world where, according to our European consumer research, two-fifths of purchases that complete on a mobile, tablet or desktop/laptop involve at least one other digital device.
A desktop or laptop computer is still the preferred digital device to make a purchase on, but consumer behaviour is changing, with mobile adoption growing substantially year-on-year. Despite strong desktop/laptop purchase preferences, more searches occur on mobile than on desktop, according to Google. Marketers shouldn’t under estimate the importance of smartphones in their customers’ purchase journeys, which, based on our research, suggests that even out of the consumers who don’t make purchases on their mobiles, half (48%) use their phone to research purchases they will later make on another device or in-store.
For the purpose of accuracy, marketers should opt for a deterministic matching solution to ensure they can properly measure the effectiveness of their campaigns and avoid any unnecessary costs associated with false sales recorded by probabilistic matching. In an industry where there are low levels of trust between agencies and brands, transparency is a must and a probabilistic solution that muddies the waters is not going to instill the trust required to work effectively with advertising partners.
Find out more about cross-device consumer behaviour.