Three Generations of Toolsets
2011: New competitors who were good at Google Ads
After years of being the (only) leaders within the market we had pioneered, we found ourselves facing two well-funded competitors each willing to pay three times as much as we did to get an order.
And these competitors knew all the tricks we had been using for Google Ads. (Don’t ask how they got this information. It’s a long story.)
Within just a few months, our orders dropped by 75% while our Google Ads cost tripled to over 50% of gross revenue.
We needed a new strategy. Quick. We knew we had a lot of Ads talent, low overhead, and some great tools. So, we lowered the retail price by 20% and went to war. How could we use Google Ads to help us find the most profitable orders?
- Isolate every possible combination of geography, product, and feature. (For example: “Anonymous STD Testing in Atlanta”. There were literally millions of such combinations to try.)
- Create separate campaigns, adgroups, and keywords to capture those combinations.
- Have a good system for measurement. Learn: what is working, what is not, and adjust. Quickly.
2012-2018: Develop and Refine Adwords Campaign Generators
Our new strategy required us to develop better Ads tools. In 2012 we closed up shop for two months while we refined our tools into a series of applications that would allow us to create thousands of combinations of geography, product (tests), and features like: “private herpes testing in Decatur Georgia”. It took us a few more months to get them working properly.
This approach worked very well. Within one month of implementation, our cost of Google Ads was back down to under 20 percent of revenue and our order volume was back to “pre-competition” levels. Extremely specific keywords and ads were a huge competitive advantage. Architecting an account to operate this way took a lot of trial and error.
In addition to the system for generating the campaigns, we also came up with a very good methodology for measuring and improving performance. Our universal measure of performance was called the “AdWords Success Index” and the process by which we determined where to spend was called the “Grid Method”. If you are interested, you can read a bit about them in the section called: “The CampaignBuilder Vision”.
By 2018, the experience and tools we had developed during this six year period proved to be very valuable. Even years later after implementation, the system that we had “prototyped” to help us sell medical testing continued to pay dividends. The tools themselves had evolved from a sequence of spreadsheets, databases, and programs to a real application with an interface a normal human might actually use.