New AdWords App That Google Won't Want You To Know About!
Posted by John-Scott Dixon on Saturday, July 4, 2009
Under: Paid Search
It's called BaddleWords, and it's probably one of my favorites because, well, I helped design it!
Here's the story, I found myself optimizing Google AdWords for a few of my clients over the last couple of years. This was a difficult transition as I had been anti-paid search for so long. I always thought that organic or natural search (Search Engine Optimization) was the most cost effective and powerful method for driving new traffic to a website. I think my bias for "lowest cost per conversion" was so extreme that it prevented me from properly evaluating Pay Per Click as a valid approach for driving traffic. I now see paid search as a technical marketer's dream. Because there is so much data available, it is a marketing laboratory. And, as in a real laboratory, you can apply the scientific method to test hypothoses.
What enabled me to accelerate my thinking was the good fortune to have a few clients that had decent budgets, and one that was spending in excess of $50,000 per month in Google AdWords. That means things happened rapidly. I was given an opportunity to perform thousands of AdWords experiments. One of the first things I began testing is the relationship between Ads and Conversion. Most of my clients had been focused on Click Through Rates (CTR) and Cost Per Click (CPC). What I want to prove was that we could determine a maximum Cost Per Conversion (based on contribution margin) that we would be willing to pay. By moving our focus to cost per conversion, we increased qualified traffic. We promoted keywords that increased conversion at a lower cost. And, ultimately, we began adjusting Ad creative to maximize conversion.
That brings us to the most interesting AdWords activity, Split Testing (aka A|B Split Testing). The purpose is to pit two Ads against each other within the same Ad Group to determine which will yield the greatest number of conversions (leads or sales). Initially, you may test different Ad titles or description lines. At some point, you'll be testing changes in punctuation. What amazed me, is that an exclamation point versus a period, where all other things remained the same, could generate a 4% lift in conversion. What became difficult and inspired BaddleWords, was tracking Ad versions. As it turns out there are a discrete number of keywords for each Ad Group - inevitably, you'll end up creating an Ad that you've already tested. BaddleWords maintains an archive of previously tested Ads, so you'll be alerted when you duplicate past effort. In addition, I created a conversion optimization algorithm with a built-in test for statistical significance. What I'm excited to report is that using this algorithm, I was able to reduce my clients' spending in half, while significantly increasing their conversion. The real opportunity, however, is to maintain budget and radically increase conversion!
Combining these elements into a single application seemed like a fun project. So, I engaged my friend Richard Stuart (Lunchbox Collective). He loved the idea and agreed to build the application with my help. Here's a few pics of my monitor while testing the application to get you excited:
Next week we'll let our first round of Beta Testers begin using BaddleWords. If you're interested in participating in the Beta test, please visit http://www.baddlewords.com to request an invitation.
Here's the story, I found myself optimizing Google AdWords for a few of my clients over the last couple of years. This was a difficult transition as I had been anti-paid search for so long. I always thought that organic or natural search (Search Engine Optimization) was the most cost effective and powerful method for driving new traffic to a website. I think my bias for "lowest cost per conversion" was so extreme that it prevented me from properly evaluating Pay Per Click as a valid approach for driving traffic. I now see paid search as a technical marketer's dream. Because there is so much data available, it is a marketing laboratory. And, as in a real laboratory, you can apply the scientific method to test hypothoses.
What enabled me to accelerate my thinking was the good fortune to have a few clients that had decent budgets, and one that was spending in excess of $50,000 per month in Google AdWords. That means things happened rapidly. I was given an opportunity to perform thousands of AdWords experiments. One of the first things I began testing is the relationship between Ads and Conversion. Most of my clients had been focused on Click Through Rates (CTR) and Cost Per Click (CPC). What I want to prove was that we could determine a maximum Cost Per Conversion (based on contribution margin) that we would be willing to pay. By moving our focus to cost per conversion, we increased qualified traffic. We promoted keywords that increased conversion at a lower cost. And, ultimately, we began adjusting Ad creative to maximize conversion.
That brings us to the most interesting AdWords activity, Split Testing (aka A|B Split Testing). The purpose is to pit two Ads against each other within the same Ad Group to determine which will yield the greatest number of conversions (leads or sales). Initially, you may test different Ad titles or description lines. At some point, you'll be testing changes in punctuation. What amazed me, is that an exclamation point versus a period, where all other things remained the same, could generate a 4% lift in conversion. What became difficult and inspired BaddleWords, was tracking Ad versions. As it turns out there are a discrete number of keywords for each Ad Group - inevitably, you'll end up creating an Ad that you've already tested. BaddleWords maintains an archive of previously tested Ads, so you'll be alerted when you duplicate past effort. In addition, I created a conversion optimization algorithm with a built-in test for statistical significance. What I'm excited to report is that using this algorithm, I was able to reduce my clients' spending in half, while significantly increasing their conversion. The real opportunity, however, is to maintain budget and radically increase conversion!
Combining these elements into a single application seemed like a fun project. So, I engaged my friend Richard Stuart (Lunchbox Collective). He loved the idea and agreed to build the application with my help. Here's a few pics of my monitor while testing the application to get you excited:
Next week we'll let our first round of Beta Testers begin using BaddleWords. If you're interested in participating in the Beta test, please visit http://www.baddlewords.com to request an invitation.
In : Paid Search
Tags: "google adwords"
blog comments powered by Disqus
I have over 16 years of experience managing and leading the Ecommerce efforts of medium and large companies. I have held sales, sales management, marketing, operations, IS/IT, legal and executive management positions in start-up to multi-billion dollar organizations. I have also served as an adjunct professor of Ecommerce for the MBA program of the University of Missouri (where I received an MBA concentrated in Direct Marketing in 1989). I led the Ecommerce initiative for Sprint PCS (PCS) and Sprint (FON) as Vice President of Ecommerce. I led the integrated marketing efforts for Insight (NSIT) as Senior Vice President of Marketing and Ecommerce. Today, I am the President of Aidan Taylor - a Web marketing company.
