expertSEM

Many tools out there push bid automation on search marketing. For some, this detached method can deliver sufficient results; oftentimes, however, bidding on autopilot simply does not work. Sadly, automation does not cover many other angles, such as syndication partner results.. With so many factors affecting success rates for search, you cannot possibly automate such a wealth of data.

I ran into this issue with a client, an online movie download provided, which I have been working on now for over four months. Naturally, the first couple of weeks are used for data collection—really striving to understand what’s working and what’s not, and gathering sufficient data that allows me to make decisions on bid strategy and other techniques. Bid strategy is always important; however, in the case of this client, the defining factor to success was not the bid strategy but that of syndication.

If you are lucky enough to have access to search data like I do, you are able to draw a report for syndication partners on all engines that you are advertising on. These reports need to include conversion data to be useful. After numerous attempts to get the conversion rate up and cost per download lower for my client via various bidding strategies, I felt the need to dig much deeper. When undergoing further analysis I noticed that there were two syndication partners pulling in significant volumes of traffic with very limited leads. These two partners contributed over 80,000 clicks and only 170 downloads. When you’re aiming for a $2.70 cost per download and the average CPC is $0.20, well, you do the math…the picture is not pretty. Most engines on the backend do discount various syndication partners; however, even at the lowest CPC, these partners still look dismal.

My first approach was to send the report to the engine requesting that they remove just those two syndication partners. After a month of fighting with them and achieving very little success, I moved onto my next strategy.

To get my point across, I decided to hit the engines where it hurt the most. I pulled syndication completely; this caused a significant crush on the daily spend. By pulling syndication I did prove my point that the results were solid without syndication ($1.81 cost per download); however, I still needed to find the right mix to keep my spend levels up and increase opportunity for downloads. So I moved onto my final, winning strategy.

I pulled the top click volume keywords out and placed them into their own “adgroup.” I then selected to push results out to syndication partners on all “adgroups” other than the top keywords. This blend worked exceptionally well. My conversion rates jumped by 3.84% overall, spend levels were up, the volume of downloads remained similar to levels prior to this testing and the cost per download is now averaging $2.81. This strategy also enabled me to work on a more effective bid strategy for the top keywords; now that they are no longer running on syndication, I am able to bid more aggressively.

So, for everyone out there that thinks you can set bid strategies then sit back and watch the profits roll in, you are sadly mistaken. Whatever the reason behind this aloof method—not enough resources, don’t understand search nuances, sheer laziness—bid automation is simply not an effective way to approach search marketing. My suggestion: take it upon yourself to manually manage your search campaigns or, if you don’t have the time or know-how, find someone who can do it for you.