Next-Level Optimization Part 2: Beyond Paid Search
Last month we took a look at identifying some ways to leverage targeting options and automated algorithms to ensure your paid search campaigns are well optimized. Now its time to up-level your optimization efforts by considering media other than paid search.
How do I know if my media mix is optimized?
If I had another dollar, in which channel would I spend it, and why?
Am I giving search marketing too much credit? Not enough credit?
At SMX East in New York there was quite a buzz about attribution management, which I though played right into my column for this month.
Last month we took a look at next-level optimization for paid search, identifying some ways to leverage targeting options and automated algorithms to ensure your paid search campaigns are well optimized. Now its time to take a look outside the friendly confines of paid search programs and see what it takes to up-level your optimization efforts by considering media other than paid search.
Don’t Panic, It’s Organic
First up is the most obvious media channel, paid search’s benevolent brother, organic search. A few months ago I took a look at a framework through which we could better understand the interaction between paid and organic search. In that post I tried to answer the specific question: If I rank #1 organically for my brand term, why should I be buying it through paid search? It turns out this approach applies beyond just brand terms.
There are many ways in which your paid search campaigns can inform your SEO, and vice versa. As an example, we sometimes comb our referral logs for organic search keywords as a source of paid search keyword growth. Conversely, if we see a paid search term driving volume with solid performance, we may choose to build content around that keyword for SEO. Going back to our previous question, you can always test the effect on your SEO of buying a head keyword using the framework we built last time around. The most important thing to remember here is to try to use a common analytics platform on which to build your analyses. Remember that different platforms will often provide very divergent data points, and thus comparing data from disparate platforms will often provide useless if not misleading data.
One Step Beyond
Outside of paid and organic search, there is a whole sub-industry developing as we speak around attribution analysis and management and media mix modeling. I wrote a column about it several months ago where we looked at a three-phased approach to attribution management. Companies are lining up to try to tackle these issues, but doing so requires a level of integration not easily achieved in the world in which I operate. There is a very fundamental issue that normally gets in the way of doing good work on this front, which is that most companies are unable to track all their media with a common technology platform.
More and more websites are beginning to see the importance of this, are taking note and making changes, but with many large companies there is still a significant analytical void to be filled. The fact remains that to conduct meaningful analyses, you need good web analytics. By “good” I mean that you should have a single source of truth for all media, built on solid technology (whether in-house or outsourced), with the ability to track all the way up to the impression level. This is normally not available with search, as the engines generally do not support third-party ad serving, but if you can get this at least for display advertising, you’re well ahead of the game.
As I mentioned in my previous column on attribution management, there are three important phases: business intelligence, statistical modeling, and actionable outcomes. And while I do believe that as a marketer your responsibility is to fully engage in every phase of this process, as someone who has been in search marketing for more than a decade I don’t recommend doing any of this on your own. There are a handful of qualified companies specializing in this business and you can bet there will be many more popping up in the near future.
But to pretend that as an above-average marketer you can effectively determine how to assign credit across media channels is absolutely irresponsible. It’s like giving me the keys to a Formula One race car and saying “well, you have a driver’s license, you can figure it out.” Don’t do it. What you should do is research vendors in this space and spend your time making sure you can put the necessary pieces together to make a tight business case internally. Ask vendor candidates for case studies that show lift in profit, revenue, or other important metrics. Get your analytics in shape. Understand your own business better. Only then can you effectively put a plan into place that will drive business success through next-level optimization.
Sell, Sell, Sell
OK, so now you have everything set up for success. Your analytics platform is robust and unified. You’ve RFPed some vendors, made a business case for attribution management, maybe you even have a vendor selected and have a purchase order open. Congratulations! Now you have one more critical step to take, and its time to put your sales hat on. In one large company where I have worked, we developed a very sophisticated attribution model in which we had a high level of confidence. The problem we then faced was that we couldn’t effectively sell it around the organization.
There were two main reasons for this. First, the complexity of the model made it difficult for people to understand—this is serious math, and frankly people aren’t going to get it. Second, and perhaps most significantly, attribution management means taking credit away from one channel and giving it to another. The scary truth is that in most scenarios we will be taking credit away from search and moving it up the stack to other events such as display impressions or email. This literally means that search (or other) marketers stand to lose credit, and therefore budget, by engaging in next-level attribution management strategies. This will inevitably cause friction and impede your progress.
What’s a marketer to do? As far as selling the complexity of the attribution model, that’s one of the reasons you’ve outsourced the project. Place the burden of proof squarely on the shoulders of the vendor—don’t bear this cross yourself. Call the engagement a “pilot” if you need to—if it doesn’t work, you can always terminate the relationship, right? As far as the other question—the part where you’re stealing budget from other (or your own) channels—the way to overcome this is to have a higher-level ally or sponsor in your organization who sits on top of all the marketing channels. If your CMO or SVP believes that attribution management will benefit the company, that may be all the air cover you’ll need.
The path to next-level optimization through attribution management is a long and winding road. Take your time pre-selling the concepts to upper management, and focus on finding the right outsource partner. If you can do that, you’ll find that navigating this tricky path isn’t so tough after all.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.
New on Search Engine Land