Smaato vs. Jumptap and Why Mobile CTR Doesn't Tell Us Much

The click is a "lazy metric." Much has been written by comScore, Nielsen and others about why clicks (or the lack of clicks) are not necessarily predictive of real-world success or conversions. Recently Nielsen issued findings that argue how little clicks correlate with sales, in an online display advertising context:

Advertisers looking to build their brands online will need to look beyond traditional web metrics to determine if their investments are paying off, according to a recent study by Nielsen. In a new report, Beyond Clicks and Impressions: Examining the Relationship Between Online Advertising and Brand Building, there is emerging evidence that brand metrics – which show attitudinal response to online campaigns – can predict offline sales. The research further shows that there’s virtually no relationship between click-through rates and brand opinion or offline sales.

(emphasis added.)

Let's be clear that CTR matters in search because you can't buy anything or find additional information without a click to a landing page. But the click has for too long been the currency of online advertising across the board. It has also been used in mobile to demonstrate "success" or performance in situations where that's clearly dubious.

In January Harris (on behalf of Pontiflex) issued findings based on survey research that said nearly 50% of mobile clicks were unintended: "47% of mobile app users say they click/tap on mobile ads more often by mistake than they do on purpose." This was research with an agenda but it does reveal the uncertainty surrounding the CTR as a mobile success metric. 

While an argument can be made that higher CTR on mobile ads is better than lower CTR, the connection to ultimate sales or success is questionable. We really need to see "secondary actions" in mobile: map lookups, calls, sign-ups, check-ins/visits, etc. to get a true sense of whether campaigns are working.

Earlier this week Smaato and JumpTap both released CTR metrics as part of larger data releases. The numbers were dutifully and largely unreflectively reported by numerous blogs and media outlets. Screen shot 2011-12-02 at 8.09.58 AM

Beyond the fact that the data disagree (compare iOS on both charts) they don't tell us much of anything. The Smaato data argue by implication that Windows and RIM are better platforms for andvertisers because of higher CTRs/response rates. However yesterday I wrote about InsightExpress findings that argued Windows users were less engaged and sophisticated in many respects than other smartphone users. 

Screen shot 2011-12-01 at 11.51.34 AM

Now compare 2009 data from ad network Chitika showing mobile OS CTRs. What can we infer from this? Android is a better ad platform than iPhone? What about Palm at the time? In fact, we can't infer much of anything, just like the charts above.

Are Windows Phone or Symbian owners better prospects or more engaged vs. owners of Android? What is the real performance of mobile advertising on these various platforms? We really can't answer that question and don't know unless or until we have visbility into "secondary actions." 

So while a CTR number may directionally indicate success or be suggestive that something is working it's not transparent enough to really declare success. Mobile offers analytics capabilities that PCs do not.

While mobile is still young the industry should learn from the mistakes of online and develop a set of alternative metrics that can be used to evaluate the true success or influence of a mobile campaign.