Often I am told that I am wrong. It seems that I hold massive misconceptions about areas of life that others find reasonable and logical. For example, up until I was a freshman in High School I thought that I could microwave food wrapped in tinfoil. I didn’t see a problem with it, and in fact I didn’t really give it a second thought. However, on the fateful day when I actually decided to microwave pizza on tinfoil instead of a plate, sparks flew out everywhere and the microwave basically caught on fire. My mom was not pleased that I ruined the microwave and I was not pleased that I ruined a perfectly good slice of pizza.
I learned that day the reason misconceptions are so dangerous is they are blind spots that your don’t even know you have. In the same way I had some misconceptions about what would happen if I put tinfoil in the microwave, many people have misconceptions about AdWords management. However, instead of merely ruining a microwave AdWords misconceptions may hold you back economically and cost you and your company money. In this blog I plan to show four common misconceptions about AdWords management so that you know that these misconceptions exist and deal with them effectively.
1. Google is On Your Side
Google is not there to help you, Google is there to help Google. People often forget that just because Google has great public relations and seems so benevolent, that they are also a business. Like any other business Google’s main purpose is to make a profit, and 96% of their profit comes from AdWords (venture beat). If you follow this logic you can see why Google would do anything to protect its precious AdWords revenue.
Here is the biggest way I see Google helping Google rather than you:
Google offers to set newcomers up on AdWords and this makes it appear like they are helping you. They are not. Google is helping Google and here is how I know: their account reps always put all of the keywords in new setups in broad match unless specifically directed not to. After seeing this happen to a friend I called the company and asked about this policy – they say they put all the keywords in broad match because they don’t want a new advertiser to miss any impressions. While it is true that you do not miss impressions with broad match – you also blow through your budget on keywords that do not match well to your business objectives. No newcomer is going to know why broad match is so bad, and even if they do understand that broad match is not great, they mistakenly trust that Google set them up to succeed. This means that the new account matches for and pays for keywords that don’t really make sense for the business.
Even worse for the new advertiser – here is what happens in AdWord’s internal system:
- As broad matched keywords only kind of match the users search, the click through rate is low.
- When the click through rate is low the quality score goes down.
- When your quality score goes down you have to pay more per click which just puts more money in Google’s pockets and less in yours.
When starting a campaign we always recommend starting with broad match modifier and phrase match and only with a small amount of initial keywords to keep the click through rate and quality score high.
2. Split Testing is Simply Changing a Button’s Color
When I began AdWords training years ago, I read pretty much every book written on the subject and they all recommended A/B testing, and the books are right – split testing is vital to AdWords success.
Nerd Note Alert – I realize many don’t know what a split test is. If you don’t know what one is, please read the following paragraph. If you already know go ahead and skip the next paragraph.
A split test is when a digital marketer has two versions of a piece of content (landing page, ad, etc) with one variable that is different. The users who encounter this content are randomly split into two groups: one group sees version A and another group sees version B of the piece of content. The only difference between the two pieces of content is the one variable that is being tested. Utilizing software to track the test a digital marketer sees which variable produces the desired behavioral change they are looking for to a confidence interval of 95%. With a confidence interval of 95%, the test is statistically significant and the difference between the two groups behavioral change didn’t happen by chance.
End of nerd note.
The common misconception is not the value of split testing, rather it is the size of what should be split tested. Books, blogs, and webinars all regale the public with stories about split testing minor changes and the massive success that come from those minor changes. I remember reading about Amazon changing a single buttons color which dramatically raised conversion rates and made Amazon millions of dollars. While that may work for Amazon, I find that those tiny tests do not work as well for smaller clients. I know from real world experience. I tried changing colors of buttons on some of my client’s pages for split testing and it didn’t move the needle at all – the reason was traffic. The more traffic, the faster a split test will conclude and the smaller the change you can make to get results. The first couple of clients we managed AdWords for had smaller amounts of traffic and it would have taken decades for the test to conclude. However, when I made major changes to the split test (whole landing page redesigns) I was able to push up the time-table for the tests to conclude and the ROI needle moved in a positive direction. If you fewer than 3,000 visitors per month I recommend split testing major changes.
3. You Can Set It And Forget It
AdWords is like those little Tamagotchi pets that were so prevalent years ago. They both are digital and both require constant attention. The difference is unlike a Tamagotchi, an AdWords account that is not getting constant attention will bleed you of all of your money. We recognize that just setting up AdWords is a time intensive process. To correctly setup and manage an AdWords account, a company has to:
- Perform audience research
- Perform keyword research
- Create ads
- Tag everything in analytics
- Create landing pages.
It often takes us 2-3 weeks just for setup alone. However there is a misconception that once the account is set up your company is good to go. You cannot set it and forget it! Not even for one day. Every single day you need to be in that account learning what works and what doesn’t for your particular audience and product. One of my favorite aspects about AdWords is that there is constantly new data streaming in. We utilize that data to perform daily and weekly management tasks such as:
- Updating bids
- Landing page reviews
- Split testing
- Adding negative keywords
- Day parting
- Checking the search terms report for new keywords
- Watching trends and more…
4. You Know Ahead of Time What Will Work
You might think you know, but you don’t. Even experts don’t know for a fact that something will work. Even with the most perfectly setup campaign imaginable – with every single best practice in the book followed, there are still surprises. Your customers won’t act like you want them to, or they use different words to describe your product than you think they should. Some of these issues can be minimized with a good strategy and research period, but until there is money on the line you just don’t know. In a research panel a customer might say they would buy something, but then when the product comes out they do not actually buy.
There can be a substantial different between a customer saying they would buy a product and actually buying a product.
This isn’t because your potential customers are mean, or they are lying. They do not even know how they would react when they encounter the product and money is on the line. Not all hope is lost though! What you want to do is closely monitor, make a hypothesis about what is going on and test in real world conditions. We have a pretty simple strategy with this: do more of what works in real world business and less of what doesn’t. If we see a certain strategy is working we learn why and do more of it, and if it is not we find out why and do less of it.