Targeting customers with cross device tracking

Devices

For brands to engage with customers, an organization needs to connect with them no matter what channel or device they are using. Easier said than done. For starters, a recent YouGov survey revealed that the average British household now owns 7.4 internet-connected devices, while OMD UK research claims that the average person, when at home, swaps between devices up to 21 times per hour.

With so many non-linear customer journeys taking place across multiple touch points, marketers need to find the best way to target customers, interact with them, and establish which campaigns have most influenced a purchase on what channel.

The use of cookies to store web-browsing data has helped to serve targeted ads for years, and while this solution works well for desktop machines, the applicability of cookies differs on the mobile device. Not just because browsers have different capabilities when it comes to collecting both first and third-party cookies, but also that user information does not always gets shared dedicated mobile apps. This, as you can imagine, leads to an even more fragmented environment.

To address these marketing headaches, we have seen the development of two methods for cross device tracking: deterministic and probabilistic.

Deterministic tracking works by requiring users to log in to an app or website. This unique identifying information (such as an email address) allows marketers to link a customer to all the devices they have logged in.

Amazon presents the most obvious example of deterministic tracking, which is able to identify where you are in your customer journey and allow you to pick it up where you left off, whether you use your desktop, tablet, phone or other smart device. Another example would be Google, which is able to synchronize the web activity of users who provide their credentials.

While an effective and highly accurate tactic for targeted ads, it has its problems. Deterministic tracking means marketing campaigns can only be conducted within these ‘walled gardens’. It also means it’s difficult to scale beyond an existing user base, or track visitors or activity outside of these circumstances.

An alternative methodology, and one championed as the future of targeted advertising, is probabilistic matching. Probabilistic tracks people by collecting large volumes non-personal device activity data, using ‘hand raising’ occurrences (an instance of logging in, registering or signing in) to ‘tag’ a device to a person. Once tagged, all previous (anonymous) history is retrospectively associated together between devices. While arguably less accurate, this approach is more scalable and enables devices to link across sites, providers and apps.

With this method, it means an office worker whose personal desktop, phone and work laptop have been linked, can be served retargeted ads at several points during his commute, working day, and when they return home.

Both approaches can be very revealing for brands and help them understand a great deal about the customer journey. For example, it could reveal customers are using smartphones for browsing and research, but their tablet or laptop to make their purchase. It can also help determine what role a paid ad, piece of content or affiliate played in the journey, rather than simply attributing the credit to the last click.

Used in conjunction with a Single Customer View, which can unify data from additional streams – such as newsletter sign-ups, email marketing campaigns, social interactions and transactional data –cross-device methodologies are helping brands to develop more rounded knowledge of their customers, with the ability to target them consistently, and engage them in real-time, no matter what device they use.


scv-cover-bv.pngWant to learn more? Read The Ultimate Guide to Single Customer View 

Whether you are a marketer getting to grips with the concept of an SCV, need to know what your organization needs to consider before building one, or want to better understand the ROI, Blue Sheep's The Ultimate Guide to Single Customer View has the answer. 

Learn more »


Topics: Article