×

Householding – The First Step Towards Cross-Device Identification

Media consumption has never been higher, with digital devices arguably the centre-point of our daily lives, which theoretically makes the case for ad-funded businesses at their height. However, the problem for advertisers is that consumption is taking place across a number of screens, making it difficult to target audience types, and calculate the ROI on media spend.

In this piece, Glen Calvert, Affectv, CEO discusses a technique called 'householding', which basically involves combining a number of clever user-identification techniques to track single users across different devices (then serve them ads).

Joining up the dots can be difficult. For years, research giants like comScore and Nielsen have struggled to provide a unified cross-device audience figure, let alone rich insights into the behaviour of individual users across the web. The advent of mobile and social has led to multiple identities, across multiple devices, and this conundrum is not made easier by the fact that mobiles change their IP address many times per day, even in the same location.

Despite its difficulties, the move towards a single-user solution remains at the forefront of the industry, and householding might just be the crucial first step.

Households

Householding provides a way to identify the devices that belong to groups of consumers who live in the same household or work in the same workplace. By identifying their location and internet access patterns we are able to infer the devices that belong to the same consumers, and the ones that belong to different consumers within the same household. This provides an obvious benefit in being able to deliver a personalised reach-frequency that tailors different messaging across different platforms. Such an approach enhances an advertiser’s ROI, conversion rates, and from a consumer point of view, delivers advertising people want, when they want it.

Devices

The development in Householding is beginning to deliver more sophisticated consumer insights beyond its primary function. Different devices do different things, and different user patterns present themselves between the home and the work environment.

Consider the example of a single user who has shown intent to travel. At work, on his desktop or laptop, we might choose to serve a message saying ‘Are you tired of working? Plan & book your next vacation with us now.’ A few hours later, that same consumer will be at home with his family, browsing his mobile or tablet. At that point it might be more appropriate to tailor the ad-text to something like: ‘Take Your Family to Florida This Year!’

The same ad, for the same brand, using the same campaign, but with a completely different message. Househoding doesn’t just allow for personalisation towards a specific user, it allows for the personalisation of messaging based on their mood and environment also.

People

As with all data – big or small – the real insights begin to flow when you apply analysis to the numbers. Two years ago, we detected a strange pattern in one of our campaigns promoting an online baby clothing store. A typical Converter Profile in such a campaign would tend to skew towards female, 25-40yrs, with a browsing history of baby apparel, women’s clothes, and food & drink (i.e mothers).

However, when we analysed the data we found that a significant percentage of our Convertor Profiles were male, 35-50yrs, with a browsing history of sports, cars, and adult sites (i.e. fathers). What we found when we mined the data more deeply was that many of our female profiles were sharing the links to purchase with their male counterparts, who were then in turn making the final purchase. By recognising this symbiosis between multiple users, we were able to evolve our targeting campaign beyond the confines of traditional demographic profiling, and produce greater conversion.

Demographic data is still an important part of online targeting, but people are individuals. Householding provides a crucial first step in the move towards collating – and applying – single user behaviour on multiple devices across the web.