There’s a light at the end of the ‘IP tunnel’ offering marketers the ability to segment when they previously had practically no insight at all. IP addresses have long been dismissed as an ineffective way of targeting consumers across the digital space.
The constant allocation, de-allocation and reallocation of devices by ISP’s together with variable registration coverage has resulted in IP being seen as the poor relation to cookie-based targeting by ad networks.
Recent years however have seen a dramatic improvement in determining the geographic location of an IP address in relation to the corresponding end user – increasingly sophisticated techniques are being used to drill down to town and even postcode level.
This has resulted in the major players within online advertising tapping into the explosion of geo-targeted ads to augment display buying. Mediamath, Turn, Videology and Infectious Media all use some form of IP address intelligence within their audience targeting solutions.
The benefits are clear: by understanding a user’s location, publishers can use ‘in-country’ knowledge (a consumer’s language, local currency and culture) to make web content relevant; advertisers can make their offers more applicable by gaining an understanding of the local area.
There are many “IP to Geo” providers (e.g. Maxmind and Quova) who collect and process publically available IP data, as well as offering cost effective solutions that can help with this high level analysis. However, where it gets really interesting for advertisers is the ability to target down to postcode level.
Companies like Digital Element, at the forefront of IP to Geo intelligence, can tap into more accurate methods of geo-location and predict IP accuracy of 5-20 miles from a consumer’s actual location. This has the potential to open up a new era of geographic targeting for brands – knowing that a consumer’s nearest car dealer is Kia or that their closest supermarket is Waitrose give rise to much greater personalisation within specific online offers.
The next step to full utilisation of the IP to postcode link, particularly within online display, is to leverage the huge amounts of offline demographic and behavioural data that has been used in traditional marketing environments for many years. Data companies like Acxiom have amassed a wealth of individual level consumer data, aggregated, weighted and modelled out to postcode or zipcode level.
These provide highly accurate predictors of behaviour giving digital marketers the ability to select on basic demographic criteria (e.g. lifestage, affluence and age) linked to postcodes that have the right mix of characteristics for a particular campaign. They can further refine these segments by the application of behavioural characteristics and offline household segmentations like Personicx. Crucially, by using postcodes (a postcode “sector” contains an average of 2,800 households in the UK), no personally identifiable information can ever be linked back to an IP address.
Knowing the characteristics of consumers within a postcode will help advertisers understand that although these individuals may well have Waitrose on their doorstep, their preference is actually for online shopping from Tesco.com; Kia may be the closest dealer, but the demographics of single professionals within the area suggest a preference for premium marques.
The key is context. Current (limited) geo-targeting techniques treat an individual as a point on the map with little, if any, knowledge of their surroundings or buying behaviour. Brand engagement can only be enhanced by this type of consumer insight at postcode level. The following statistics published by Digital Element speak for themselves:
- 30 – 300% increases in response rates to geo targeted content
- Web visitors are 6 x more likely to convert when content is localised
- Geotargeted impressions sell for 30 – 50% premium over untargeted
In conclusion, IP address geo-targeting, while not completely solving the complexities around linking offline and online channels, goes some of the way there and gives a real boost to marketing performance when the marketer knows practically nothing else about the individual. Advertisers can find the right audiences at scale while sidestepping the tricky issues of cookies, PII and privacy: In fact, it could be argued that ads placed in broad geographic context with relevant content, are more effective at engaging consumers than those personalised on recent search or navigational behaviours.
