Build out Keywords
Before you can do anything, the first step is to figure out what you are going to bid on. With no historical performance data available, you’ll need to pay for information to find out what works and what doesn’t, so adding thousands of keywords off the bat can be an expensive and ill-advised plan.
It makes a lot more sense to initially focus your efforts on smaller groups of keywords that are more manageable and likely to perform well. Intent-related and product-specific terms are probably going to be best, so spend your time on those. If we’re looking to sell lawn ornaments, be aggressive on ‘buy lawn ornaments’ or ‘cheap garden gnome.’
The AdWords keyword tool can be useful for coming up with ideas and expanding on your initial build, but you’ll need to go through the sets carefully and pick out keywords specific to your product or business – ‘flamingo lawn ornaments’ is a great keyword (if you sell those), while ‘gardening’ probably isn’t.
Traffic estimates are also useful but need to be taken with a grain of salt. Google consistently over-estimates on rank while under-estimating CPC. Not being able to see broad match modified traffic numbers also hurts the usefulness of this tool, but it’s a good place to start.
Keep it Simple
When there’s so much to do, it’s extremely important to make sure that you’re prioritizing tasks correctly and not getting caught up in details/projects that are time-consuming but won’t show large performance improvements.
Setting up geo splits and dayparting strategies are great examples of this. You will of course need to set location targeting and dayparting if you’re only selling in certain areas at certain times; beyond that, though, if you’re just getting going it’s best not to worry about geo segmentation or hourly bid modifiers. Look into this once you have actual data. In the meantime, prioritize working on ad messaging and make sure you have sitelinks up from day one. Sitelinks are awesome.
We can get an idea of traffic volume using the traffic estimator tool and ballpark conversion rates by taking a look at direct or organic numbers if they’re available, but you won’t really be able to model performance accurately without any data. As a result, it’s best to take a conservative approach. Daily budgets are a safety net that absolutely need to be put in place to make sure that spend doesn’t dramatically exceed expectations. Vanilla broad match without negatives can also result in some unpleasant surprises, so try to experiment with broad match first. Above all, just check in on the numbers at least a few times throughout the day to make sure everything is looking good.
Pre-launch, setting up ad groups with relevant keywords, ads, and landing pages is the main goal. Once you get going is when it really gets interesting. No campaign rolls out optimized from the beginning, but the great thing about search is that you’re able to adjust and tweak so many things to drive improvement.
Look for part 2 of this series which will discuss how to optimize your account after the first wave of data rolls in from the launch.
About the Author
Eric Smith joined PPC Associates in September 2012. He has been running enterprise scale SEM accounts since 2011, specializing primarily in e-commerce. Eric graduated from the University of California, Berkeley with a B.S. in Microbial Biology. Prior to his entry into digital marketing, he worked in a research laboratory. In his free time he likes to swim, read, study language, and travel.