There has been a lot of movement at our office over the last and current week, a lot of colleagues coming in to finalize the details of our product launch. Given the nature of our product launch, I cannot write much about it, but I can tell you that in about 20 days I will be able to give more details.
In the meantime, let’s talk about one of my assignments: to find the top 10 Google AdWords for us. The main goal is to set up a monitoring systems that allows us to monitor these keywords in conversations relevant to us on Blogs, Twitter, Friendfeed, etc. Sounds easy? I wish!
Finding the right (key)words is hard, just like NY prepsters, Vampire Weekend sing on “Oxford Comma”:
Haven’t got the words for you / All your diction dripping with disdain / Through the pain
So what is a web analytics newbie to do? I set on the following 3 tasks:
- Review Brian Clifton’s “Advanced Web Metrics with Google Analytics” for all mentions of Google AdWords.
- Review Google AdWords’ Keyword Tool.
- Read relevant blogs on the subject.
Given the urgency of our product launch, I have only been able to do the two first tasks so far (and have not been able to blog about it, until today!).
Let’s see what I got:
First, the book by Clifton is really great and I can see referring to it a lot in the future. It already has lots of page separator’s by subject. If this book is either sitting on your shelf or on your Amazon.com wish list, here is a motivator to pick it up. Check the following pages on tips regarding AdWords: 73-75, 92, 103-105, 173, 190-201, 299-305 and 308. The tips range from setting objectives of your AdWords campaign to creating filters on Google Analytics reports to determining key measures such as cost per acquisition.
One of the main problems with selecting and monitoring keywords is the specific/broad dilemma, that is whether to choose a campaign based on “shoes” versus “blue shoes”. A broad match set on “shoes” would allow you to measure web visitors that searched for “blue shoes”, “nice shoes”, etc., while a specific match set just on “shoes” would only measure web visitors that searched “shoes” only. Also, a specific match set on “blue shoes” would not include a keyword search like “pair of blue shoes”.
Second, after reviewing the Google Adwords’ Keyword Tool, the mentioned problem becomes evident. By generating keywords using descriptive words or phrases I get some interesting results. For example, I am particularly interested in “photo sharing”, which has an approximate average search volume (AASV) of 165,000. Not too bad…I thought. Just the keyword “photo” has an AASV of 20,400,000! The keyword sharing has an AASV of 2,240,000! What is a web analyzer to do?
Think outside of the box!, said Vampire Weekend.
So if there’s any other way / To spell the word / It’s fine with me, with me
The next step is to generate keywords using your actual website content, let it talk to you. After inputing our company address I got very good leads like “photo gallery” with a healthy AASV of 823,000 and “slideshow” with 1,000,000. Funny, these are key features that I had not consider before and they make more sense (besides having a greater AASV!).
- Check your handbook / It’s no trick: There is a lot of literature available on selection of keywords. Try the tutorials at Google AdWords and AdSense first, then complement your ideas with Clifton and web analytics’ blogs (refer to my Blogroll on the left side for some references).
- Adjust my tie / Know your butler, unlike other guys: Two words > Keyword Tool! Listen to your website. Let it do the legwork for you.
- I met the highest lama / His accent sounded fine: Quality over quantity. Selecting a single word with a high AASV may provide you a greater hit rate, but these site visitors may just exit on your home page after realizing that your site has nothing to do with their search. Remember the key is conversion!
Vampire Weekend playing "Oxford Comma" at Pipeline Cafe (Honolulu, HI). Photo credit: me!