Phocuswright LA – Cross Device Marketing

Solving for Cross-Device Marketing

At the recent Phocuswright Conference in LA, Daniele Beccari- Head of Travel Products at Criteo discussed the difficulties faced by advertisers in cross device marketing, and targeting and tracking today’s travelers as they plan and book their trips across multiple channels and devices.

Conversion rate is a key metric on travel websites and it is often not measured correctly. The issue is that many users use multiple devices to research and buy their trips online.

  • Mobile smart phones
  • Home computers
  • Laptops
  • Work computers
  • iPads/tablets

Consider the traveler who starts to research their trip on their mobile phone in a coffee shop, before doing some additional research on their work computer.  They continue to narrow down their options using their laptop at home before finally booking on their iPad in bed that night. Obviously this path to booking does not all necessarily take place on one day but nonetheless, the booking journey is spread across 4 devices and 4 different traditional tracking cookies.

The limitations of cookie tracking means that this single user is being tracked separately across these devices. The final booking on the iPad can appear as a ‘magic booking’ where a visitor appears to book on their first tracked visit to the site. It’s not actually their first visit to the site, it’s just their first visit on that device, therefore a new cookie.

cross device marketing

 What problems does this cause for marketers?

This behaviour and the limitations of cookie tracking is a big problem for marketers trying to target potential customers with ads and to analyse website behaviour.

Conversion metrics are undoubtedly flawed but the issue if bigger that, attribution models and A/B testing are also affected. The same user can appear in both the A and the B cookie pool on different devices.

What’s the solution?

The behaviour of using multiple devices on the travelers path to purchase is here to stay, so what solutions are available?

Probabilistic Match

A probabilistic match is an educated guess, and can be based in I.P. address, usually in combination with user behaviour such as visiting at similar times and behaving in a similar way. Marketers try to encourage users to create an account and log in. This helps marketers to track users across devices. For this to work well it is important for marketers to incentive account creation and logging in.  It’s also important that this is an easy process for the customer. Offering a ‘keep me logged in’ option as an example of good user experience.

Exact Match

Exact match tracking is possible when users are ‘logged in’ to the same account across multiple devices, for example to Facebook or Google. Exact match can also work where users use the same email address to log into multiple accounts, like Facebook, Twitter, Travel apps etc. The email address is ‘hashed’ using a mathematical formula to encrypt it. This encryption creates a cookie that is unique to that email address, and assigns the same encryption when logged into multiple accounts across multiple devices. This makes cross device tracking infinitely more accurate.

Privacy Concerns

This type of tracking obviously comes with privacy concerns. A key to privacy is transparency. According to new privacy rules, when a user opts out of allowing once of these cookies on one of their devices, the user must be opted out on all devices for that particular cookie. This makes sense for the visitor, as otherwise it would become too unwieldy for people to manage their privacy online.

TIP: Facebook allows users to check how may devices you have logged into Facebook on recently. Check it out – the number might surprise you, it’s in the security section of your account!

What we have learned

The key takeaways from this presentation were that a cross device journey to purchase is here to stay for travel companies.

Implementing effective cross device tracking is becoming increasingly important for marketers, this is vital to improve the accuracy of conversion tracking, A/B testing and attribution.

Privacy is of vital importance. When choosing a vendor for this type of tracking ensure that they offer the functionality that opts users out across devices when they opt out on one device.

For more information on Daniele Beccari’s presentation please see