Google began rolling out a huge update to its device bidding in Q3 2016, as advertisers became allowed to set separate desktop and tablet bids for the first time since the migration to Enhanced Campaigns in 2013.
While tablet performance for most advertisers wasn’t as good as desktop even when Enhanced Campaigns launched, the gap between tablet and desktop performance grew over the years, and Google’s argument that these two device types should be lumped together for their similarities steadily lost credibility. As such, the 2016 update was cheered by most marketers, who had long bemoaned the lack of control afforded in the Enhanced Campaigns set up.
But are brands really taking advantage of this new control like they should? Here we’ll show how separate desktop and tablet bids have impacted Merkle data, as well as share evidence of what we believe might show that a lot of advertisers aren’t quite flexing these new bidding options like they should be.
Moving Tablet Cost-Per-Click in Line with Revenue-Per-Click
The big reason we wanted greater control over tablet is that the revenue-per-click (RPC) of tablet devices was steadily getting worse and worse relative to desktop, down to about 35% lower in Q2 2016.
Without bidding controls, however, advertisers were forced to bid the same for tablets as desktop, resulting in only 5% lower cost-per-click (CPC) in Q2 2016.
Looking at how relative cost-per-click (CPC) and revenue-per-click (RPC) have moved since Google’s Q3 update, tablet return on ad spend is now nearly identical to desktop. Not only has relative CPC declined with advertisers making bidding adjustments, relative RPC also increased as the worst performing keywords on tablet have been bid down out of the results.
This significantly affected overall trends, sending desktop spend up with an increase in CPC and RPC and tablet spend down with declines in traffic and CPC.
While we predicted that desktop CPC would go up as advertisers shook off tablet performance from bid calculations, what’s been a bit more surprising is how big the improvement in desktop click growth has been. After desktop paid search traffic declined 10% Y/Y in Q2 2016, click growth has since rebounded and sat at positive six percent growth in Q1 2017.
If all advertisers were bidding up on desktop in unison with the new bidding controls, it seems that desktop traffic gains shouldn’t be this big as everyone would just end up paying slightly more and not moving in terms of average position. But what if all advertisers aren’t bidding desktop up with the new controls?
Merkle Desktop Average Position on the Rise
For the median advertiser at Merkle, average position moved up the page significantly beginning in Q4 2016 (remember, negative change in average position indicates movement up the page). While the jump in average position in early 2016 can be attributed to the removal of right rail text ads at the same time, the additional movement which began in Q4 aligns with our observed increase in desktop click growth.
This suggests that our ads are pushing past other ads on the page as we bid up desktop using the new controls. Thus, we think that other advertisers out there might not be bidding up desktop as aggressively as the value of the traffic might warrant.
As such, it’s possible that Merkle advertisers are benefiting from early adoption and implementation of separate desktop and tablet controls.
Most bidding systems, including Merkle’s own proprietary bidding, required significant changes in order to fully take advantage of the new bidding controls, thus the small delay between Google rolling out the changes in Q3 and our own data reflecting the new capabilities in Q4. It’s probably safe to assume that all brands will eventually catch up in taking advantage of the changes, but for now there appears to be an early-mover advantage.
In short, the uncoupling of desktop and tablet has helped Merkle better allocate spend by device type, in turn helping to accelerate growth as waste was reduced. Further, our advertisers appear to be potentially benefitting from competitors failing to bid up as aggressively as we are, pulling in more clicks as average position moves up the page.
The takeaway, then, is simple – take advantage of desktop and tablet bidding to better optimize your campaigns. If that’s not possible using your current bidding tools or campaign management, then try to figure out how to make it a possibility moving forward. It’s certainly helped our brands get more out of paid search and will likely do the same for you.
Or don’t, and let the early movers keep gobbling up all the click growth on desktop while tablet performance keeps weighing on your desktop bids. It’s really your call.