Traffick - The Business of Search Engines & Web Portals
Blog Categories (aka Tags) Archive of Traffick Articles Our Internet Marketing Consulting Services Contact the Traffickers Traffick RSS Feed

“Concrete Evidence” On Tablet ROI in AdWords (Exhibit A)

Posted September 20th, 2013 by Andrew Goodman

At a recent conference, a Google executive reiterated what has become a surprisingly sustained refrain from Google since the full switchover to the Enhanced Campaigns architecture in AdWords: Google believes that “tablet behavior is about the same as computers,” and that there is “no concrete evidence” that any disparities in behavior warrant separate bidding capability for this device type. Yet Google feels that “this does not mean that Google plans to take away the separate reporting, as Google has always maintained the principle of providing data and not shutting off data breakdowns that can help advertisers gain insight.” Do you feel lucky?

Most PPC advertisers are now well aware that all of the granular bidding options for various mobile devices (including OS, device types, etc.) have been taken away under Enhanced Campaigns. Some of this is available for Display, but much of the control was removed for Search. Despite the removal of a “mobile only” campaign capability, some of us believe that the bid factors for smartphones are a convenient way of using a heuristic to simplify account management. The smartphone bid capability is now customizable to the ad group, allowing us to avoid wasting funds on a channel that needs a different bid structure to be profitable for most advertisers.

So the question remains: when will Google let us control our tablet bids? Want to see “evidence” that we urgently need to stop wasting funds in this channel?

All we have to go by is our clients. I’ll start with data from one dart-picked client, and keep going until I get farther down the alphabet. I will also be holding my breath until I turn blue.

In order to maintain client confidentiality, I’ll refrain from sharing date ranges and other private data, but the date ranges are long & the data is significant. The industry segment may be disguised, also.

Client A

  • In a printing-related business
  • Success is measured by e-commerce transactions (sales) and revenues (CPA, ROAS).
  • Average CPC is identical between tablets and computers; tablet CTR’s slightly higher.
  • 7.7% of spend is on tablets; less than 1% is on smartphones as we’ve chosen to largely avoid the latter
  • CPA on tablets is 5.4X that on computers (440% higher)
  • Tablet ROAS is 80% worse than computers (or, it’s 20% the computer number).
  • This is on search campaigns. In our Remarketing campaigns in the Display Network, we have not a single conversion from tablets, ever. Depending on the month, this channel amounts to 10-15% of remarketing spend, but we can’t shut it off or even turn down the bid, despite it never having converted.

Of course, people may switch devices and attribution may not be perfect. And some advertisers may not have a website that makes it quite so hard to customize a purchase on some tablet OS’s, as this advertiser’s does.

But it should be up to the advertiser to decide how to handle bidding on all of their significant, identifiable segments.

Exhibit B… to follow soon!



5 Responses to ““Concrete Evidence” On Tablet ROI in AdWords (Exhibit A)”

  1. Bill Laidlaw says:

    One tablet (iPad) conversion for your blog Andrew:)

  2. Andrew G says:

    Yep, people definitely read voraciously on tablets. :)

    Unfortunately they sometimes don’t complete complex ecommerce transactions on tablets or smartphones, and maybe never intend to. Bid control is king!

  3. Metz says:

    I agree to this Andrew, “But it should be up to the advertiser to decide how to handle bidding on all of their significant, identifiable segments.” And I agree about the ecommerce, providing everything needed to sell online that you mention under Client A. Great article!

  4. Larry Kim says:

    how about optimizing the tablet experience for the client?
    that share of tablet search is just going to get bigger and bigger over time.

  5. Hi Larry, sounds like a little bit of victim-blaming there.

    Of course, websites should work well on tablets. Not all do. But for some industries, intent on tablets is simply way different.

    Moreover, the tablet ecosystem of late has actually been fairly fast-moving. Not every company can roll out new platforms every few months. Some companies, believe it or not, are caught up in international organizations that centralize responsibility for IT decisions. Some companies aren’t big enough to spend enough on website changes often enough to be nimble enough to keep up with behavior.

    *Simply* saying that the tablet experience should work better without regard to the fact that in the *meantime*, a company should have the flexibility to change bids on that traffic, is a bit like saying to a company who is out of stock on an item that they can’t pause their ads, so they should “just hurry up and get it in stock”!


 


Traffick - The Business of Search Engines & Web Portals

 


Home | Categories | Archive | About Us | Internet Marketing Consulting | Contact Us
© 1999 - 2013, Traffick.com. All Rights Reserved