There are two main ways of looking at the value of branded paid search:

1. It does not make sense to run branded search if you have strong organic rankings for your brand because your organic listings will pick up that traffic and conversion. Why pay for something that you can get for free?
2. People who search for a brand term are often in a research mindset rather than a purchasing mindset, and with paid search ads you can present these potential customers with a much more compelling offer and land them on the ideally optimized page to convert traffic. Your conversion rate and total site conversions will be higher if you run a well executed brand paid search campaign. Also, if you don’t bid on brand terms than competitors who bid on your brand terms are more likely to steal traffic and revenue from you.

For the test we are trying to figure out how to maximize our paid search spend in terms of generating total conversions. We have seen that our non-brand search spend has a tremendous impact on the organic traffic and revenue of this particular client. We don’t care who gets credit for the conversion that comes through; we are only concerned about maximizing conversions as a whole. Therefore, our goals for the brand test are to:

• Understand whether we should be running branded paid search terms
• Figure out how to truly maximize site bookings for a given budget
• Better understand the relationship between paid search and natural traffic and conversions

To hopefully accomplish these goals we are running a period of true non-brand search (with the true brand term set as a negative keyword) followed by a period of full branded paid search. We plan to keep spend levels perfectly constant between the two periods and monitor total conversions through Omniture. We are going to ignore Bluestreak paid search data entirely and look at combined organic and paid conversions. The period with higher conversions wins.

Of course, there are many variables that can throw off our results: holidays, conversion window, organic search trends, paid search competitors, etc. There is no way that we can run a perfect test. Our hope is that something will really jump out at us though and we can feel confident coming to some conclusion.

Right now, we are in the middle of our test. We are running just non-brand. In a few weeks we are going to run full force brand and compare results between the two periods. We are eager to see how the test turns out and get back to the client with our recommendation as to whether or not they should be running branded paid search.