I have long been a fan of bid management solutions. Back in the old pre-Panama days, I made a bundle of dough riding bid management though Yahoo! Search Marketing. Before we implemented bid management, we were spending about $10K per month on PPC and barely breaking even. Once we implemented bid management, the margin rapidly climbed to 30 percent.
Now that was interesting, so we began to invest more energy into those PPC campaigns. We scaled the spend many, many, many times over our original spend, and kept the margin at a nifty 30 percent. Life was good!
But there may be a better solution available now. In today's column, I will discuss the problems with traditional bid management solutions, and a case study involving Travelzoo and Efficient Frontier. This case study will show how portfolio based bid management brought positive results to Travelzoo.
The Problems with Rules Based Bid Management
With the advent of Panama, Yahoo! now has a PPC product that is extremely similar to Google's. Old fashioned bid management tools just aren't quite as effective in that type of environment. This is largely because traditional bid management solutions use a “rules based” approach.
What this means is that it attempts to evaluate the ROI on a keyword-by-keyword basis. This sounds sexy, but ultimately, this approach has a difficult time working. Think about it for a minute. If you have (only) 1,000 keywords in your campaign, how many of those keywords have enough click-through traffic that you would have enough data to perform a meaningful ROI calculation? How many clicks would you need? 20? 50? Even though 1,000 keywords is not a huge campaign, with the above number of keywords, you are already chasing the long tail terms that are low in volume. You may never have enough volume to make a decision on over 95 percent of your keywords.
In addition, this type of system does not offer the potential to make tradeoffs between the spend on one keyword versus another. As we will show, this will at times result in a lower total ROI.
The Theory of Portfolio Based Bid Management
In portfolio based bid management, you do not set keyword-by-keyword goals for your campaign. You provide a much simpler input in the form of business goals for the entire campaign. Efficient Frontier performs this in three major steps:
- The first step is to get some keyword and conversion data in the market place. Efficient Frontier obtains data from the search engines, and from its customer's accounts. It also obtains information from various sources on which keywords are related to other keywords. This information is used to put the keywords into logical groupings, so that incoming conversion data can be reviewed across a set of keywords.
- The second step is to put together a forecast based on the campaign. This is done by Efficient Frontier at the ad level.
- The third step is to feed this data into a linear programming solver that manages the campaign and optimizes overall campaign results. The linear programming solver sets up initial bids based on the forecast and business goals, monitors the results, and rapidly adapts as new data comes in.
Travelzoo Case Study
Travelzoo publishes travel offers from over 600 advertisers, sending the information to over 11 million subscribers in the United States, Canada, Germany, and the United Kingdom. Here are some quick facts about the makeup of its subscribers:
- 23% of its travelers make 5+ trips per year
- 73% have a valid passport, even though the U.S. average is only 20%
- 75% have a college degree
- 34% have a household income above $100K per year
Travelzoo had set goals of increasing conversions by 15 percent and reducing costs by 25 percent. In fact, the results far outstripped these goals. Here are some of the results it obtained:
- Increased sales by 27%
- Increased profit margin by 64%
- Lowered cost-per-action by 6%
Not too shabby! Better still, initial positive returns took only a few weeks to materialize, making the whole campaign easy for Travelzoo.