Which Metrics Should Be the Focus of Your Mobile Campaigns?

Advertisers are seeing rapid growth in their mobile traffic driven so it's natural that mobile advertising is growing apace. But it's still an area in which marketers and agencies are finding their feet. Are the metrics that are often applied to traditional online advertising served on desktops and laptops - such as CTR - valid for the smaller screens of smartphones and tablets? The conclusions of a new report conducted by Nielsen on behalf of global location marketplace xAd would suggest they're not.The study, which was conducted over Q1 and Q2 2014 included 12 national brand in the US running 80 individual campaigns which delivered 200m impressions. Three different approaches were taken to attribution - CTR, Secondary Action Rate (SAR) including actions such as click-to-call, looking up directions or clicking for more information, and Store Visitation Lift (SVL), defined as the %age of the -exposed audience that visited the advertiser's location divided by the %age of the non-ad-exposed audience that visited advertiser.The conclusions were that using CTR on its own was not a good indicator of ad engagement or awareness. In fact, it was the exact opposite. The study found that when CTRs were higher, Secondary Action Rate (SAR) tend to be lower. Worse still, when ads were optimised to CTR alone, the adverse correlation become more striking. Looking at the ads in the survey in the retail sector, optimising ads to improve CTR by 40% decreased SAR by 69%. In the automotive sector, a 16% improvement in CTR delivered a 46% fall in SARs.When compared to SVL, the same inverse correlation was found - the lower CTRs wereoften associated with the highest offline in-store visitation rates.Optimising campaigns around SARs, however, had more of a positive impact on the optimised actions, and less of a negative impact on CTRs. So a 219% increase in SARs for retail advertisers only caused CTR to fall by 27%. The figures for the automotive sector were +86% and -14%.So what does this tell us? At best, campaigns should only be optimised around CTR if driving top of the funnel awareness is the goal. It's clearly not a true indicator or interest or desire otherwise it wouldn't have a negative correlation with lower funnel activities like SAR and SVL.In truth, though, this study discredits CTR as a viable metric in mobile advertising - just as it has been discredited in online advertising served on desktops and laptops. Previous non-mobile studies have shown that CTR has little or non-correlation with purchase intent, and with a greater chance of accidental clicking on mobile devices, it's clearly an even less reliable guide when it comes to mobile campaigns.As a replacement, neither SAR or SVL are perfect. SAR works better for shorter lead time, less considered purchase where advertising is more likely to lead to instantaneous actions. SVL is the best measure of purchase intent - as it measures driving actual footfall into stores - but will work best for brands with high levels of offline purchase and with lower levels of footfall (consistently high levels of footfall will make campaign related uplifts difficult to observe.)To optimise mobile campaigns, brands will have to chose the right metric(s) based on the campaign's goal and their industry sector - and a weighted hybrid model is likely to be the best way to go in most eventualities. But it's highly unlikely that CTR is going to play any part in that model - it simply doesn't make the grade for mobile marketing.