Sign up for our daily recaps of the ever-changing search marketing landscape.
Paid Search Advertising Gets More Complicated In 2013
Last week, we assembled a group of paid search experts in NYC at the SMX East Conference for a conversation about where Paid Search Advertising was heading in 2013.
I figured we’d hear some wild predictions about new mergers and buyouts, ad formats, campaign automation tools; and the usual whining about rising CPCs, Google’s quality score and ad serving policies; and whether or not Microsoft would make a significant dent in Google’s search marketshare.
Instead, our panel, which included some of our industry’s most experienced paid search luminaries like Chi-Chao Chang of AdX, Kevin Lee of Didit, Kimm Lincoln of Nebo, Craig MacDonald of Microsoft, and Sid Shah of Adobe talked more about fundamental ways that search behavior is changing as a result of new devices.
I’ve tried to capture some of their thoughts on where paid search is heading next year to give you some food for thought as you look ahead to plan and budget for paid search 2013.
Desktops, Tablets & Mobile Phones
Our conversation started on the topic of the huge increases of non-desktop devices that people use for search. Sid Shah believes, and other panelists agreed, that mobile search clicks may account for as many as 20-25% of all search clicks in 2013.
I think that most paid search managers know by now that you can target mobile phones, tablets and desktop computers distinctly in Google AdWords and Bing Ads campaigns, and that at minimum, you should target the mobile phones separately, so that you can bid them appropriately (usually lower bids) and direct users to mobile-friendly landing pages.
The question came up, “Does it make sense to target tablets separately, too, or just lump them together with desktop computers?”
Microsoft’s Craig MacDonald suggested that while mobile devices contribute incremental clicks and have their own specific use cases, tablets and desktop searches are “very, very similar” in spite of media reports to the contrary.
Kevin Lee and Chi-Chao Chang, on the other hand, argued that search behavior on tablets can be markedly different from desktops because of how, when and where people are using these devices. Tablet usage typically spikes in the evenings and weekends, when people are reading e-books, watching TV with their iPads on their laps, or engaging in other leisure and social activities.
Although tablet-based search queries may indeed be very similar desktop queries, the queries themselves take on new meanings and implied intents depending on the place, time-of-day, and the environment in which the tablet is being used. For those reasons, Lee and Chang thought that separating tablets into their own campaigns makes a lot of sense.
AJ reported that for the first time ever, his analysis of comScore data shows a a year-over-year decrease in U.S. desktop searches across Google, Yahoo and Bing, as illustrated below:
The implication, of course, is that if you haven’t already started to think through your strategy for addressing the changing shape of the search device landscape, now would be a very good time to start.
Increased Complexity In Your Accounts
More targeting, of course, means more complexity in your accounts.
Since device targeting in Google AdWords is done at the campaign level, that means that you need to consider device-specific campaigns for each of your other targeting segments, such as display, geography, brand, non-brand and so on.
Where you may have had 50 campaigns before, you’ll need 150 to cover mobile, tablet and desktop. The same holds true on Bing Ads, except that you have the option to target devices either at ad group or campaign level.
As you can see, targeting can increase the complexity of your account very quickly and overwhelm your ability to manage it — either manually or with automation tools. My colleague and fellow Search Engine Land author, Benjamin Vigneron wrote an excellent post on managing account granularity versus scalability last week that is definitely worth reading.
Target People, Not Keywords, Devices & Locations
The true complexity of paid search in 2013, however, won’t just come from the cut and paste mechanics of creating and then managing multiple campaigns, but rather from the extra effort you’ll need to apply to creating and testing new ads and landing pages as you optimize your new targeting campaigns.
Craig MacDonald told us to expect search engine targeting to keep evolving in 2013 from simple keyword, device, and location filters, to include more social and demographic data filters, too. Both Google and Microsoft are working on the ability to allow users to do real time bidding on search keywords that take recent search history into account.
For example, if a particular person searches on [automobile] you may want to bid up when they search on word [auto insurance] in order to drive improved conversion rates.
There is a flip side to this progression of ever-improved targeting, too. As search engines give us more social and demographic targeting signals, it then becomes our responsibility to define our audience targeting segments.
What that means, of course, is that we actually have to know who our customers really are. We need to know how old they are, what professions they work in, what they read and what they do on weekends. Instead of knowing our keyword inventory and match types by heart, we need to know what is in the hearts of our actual customers.
Kimm Lincoln referred to this as “human-centered PPC” thinking. Instead of thinking only about keywords and ads, we need to also approach our campaigns from our customers’ perspectives.
In other words, do more of what traditional offline marketers have always done. Kimm suggests we incorporate insights from the rich demographic and psychographic data we get from online social networks on Facebook, LinkedIn, and Pinterest, as well from offline networks like sales reps, user conferences and so on, into our paid search program designs right from the start.
Tracking & Attribution
Tracking and attribution, unfortunately, may not get better in 2013, and may even get worse. As Sid Shah put it, not being able to track a conversion across devices is a huge blind spot.
Between Google removing more search queries from query strings and the ongoing challenges of trying to understand user behavior as as they switch back and forth between their phones, tablets and desktops, our ability to correctly attribute conversions will take a few steps backwards in 2013.
Bottom Line For Paid Search In 2013
The bottom line for search marketers is that mobile phones and tablet devices are significantly transforming the ways consumers search, make transactions and interact with brands online. Search marketers and many website owners are simply not keeping pace with these foundational changes.
The most important investment that most companies can make to improve (or maintain) the performance of their paid search campaigns is probably going to be in mobile infrastructure, U/X engineering and design.
Few companies have developed truly awesome mobile websites and landing pages that address the new demands and expectations that consumers have for immediacy and urgency. There is a big difference between a site that has been developed with the user and device in mind, versus a desktop site that has been converted, or stripped down to be a ‘mobile’ site.
The second area where you should consider investing is in training. Specifically, invest in training your paid search marketing team on traditional marketing concepts and applications. In many companies, PPC managers are generally more technically oriented than consumer savvy. They are probably more comfortable manipulating keyword lists, Excel spreadsheets and analytics reports than they are with describing and defining consumer behavior and motivational triggers.
Deepening your team’s skill set and changing their process orientation from from an inside-out, keyword-centric approach to one that incorporates a more holistic human-centered thinking will pay dividends in 2013 and beyond.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.