The last-click has been the unchallenged method for measuring how customers enter the sales funnel for a long time. However, last click models don't provide a comprehensive picture because the path to conversion is never linear, it's multi-channel and multi-device.
Consumers navigate many touch points, both online and offline and it is impossible to trace just one definitive trigger point for a sale, which in effect, is what last click does. This complex reality calls for a more sophisticated measurement solution.
Improving attribution is an industry work in progress and there are limitations, primarily because every advertiser is different and has different objectives which cannot be reduced into a "one-size fits all" attribution model.
This means a consistent standard for the actual methodology of attribution is highly unlikely.
Where marketers can and should be striving for more effective attribution is by seeking a unified view of their customers to better understand the consumer journey.
Companies which have a high percentage of logged in users have a better grasp of cross-screen analysis and how customers move between their various devices.
A key problem with last-click attribution is how much it undervalues display advertising, and therefore prospecting. Retargeting naturally yields a better clicks-per-ad ratio, meaning that as long as last-click is the measure of success – and media planners use it to allocate spend – there will be a continued skew towards click-heavy retargeting campaigns.
According to our research, the last-click model is currently undervaluing display by as much as 150 per cent.
The knock-on effect is a growing emphasis on existing customers and less on reaching new audiences.
Advertising shouldn't just focus on reminding consumers to come back to a store after they have left it, but as long as brands and advertisers prioritise last click this is in effect what they're doing, thereby limiting the size of their potential customer base.
Viewability is an issue that sits at the heart of any potential attribution model. If brands really want to appreciate the value of the media they've been buying, they have to know that the ad was seen by the user.
Fortunately, online advertising is becoming more and more sophisticated, helping advertisers to avoid the problems associated with practices like cookie-bombing. Yahoo's native advertising, for example, only renders as a user scrolls, thereby avoiding multiple below-the-fold ads.
Native is definitely the direction in which advertisers should be looking. According to our previous research into mobile native advertising, smartphone users are more than twice as likely as other consumers, to agree that they don't care if the content they read is an advert, provided it's engaging.
This means greater – and more resonant – brand interaction through mobile devices, particularly as we mobile native ads deliver 72 per cent more emotional response than other types of 'in-stream' advertising.
The holy grail of attribution, and where it will be become very interesting in the coming years, is finding a holistic view of online and offline.
The mobile phone is will be integral to this, taking into account, for example, consumers' location based interactions with an outdoor billboard.
In the meantime, attribution remains a difficult and complex topic and we're not at a point where a standard has emerged.
However with brands continuing to ask questions, and media and data owners continuing to push for a better solution, we'll soon see the technology progress and case studies emerge that will move the conversation forward.
Dora Michail, senior director for audience solutions at Yahoo EMEA