Why Paid Search Campaigns Often Fall Short of their Volume Targets

Why Paid Search Campaigns
Often Fall Short of their Volume Targets White Paper June 2008
One of the big challenges in search engine marketing is getting all of the volume you want from your campaign. For example, if you know that you are willing to pay $10 for a conversion, and your conversion rate is 10%, you would most likely be willing to pay $1.00 for each click to get the highest number of conversions you can. However, the available volume in paid search is sometimes limited, and prevents you from getting to where you want to in your campaign. In other words, people only search so much on keywords related to your campaign, and there is nothing you can do with your paid search campaign that can change that. This is one of the major limitations in search engine marketing today. Large brand Marketing VPs, for example, might find the cost per conversion metrics very compelling, but if it does not scale to a level that produces a significant number of new customers, it just does not make the radar screen. Put another way, if your ROI on your paid search campaign is 10 times higher than what you get on your other marketing campaigns, but produces only 1% of your new customers, who cares? This very same factor is a limitation on the growth of smaller businesses as well. Getting at as much (profitable) volume as you can is what it ends up being all about. Hitting your CPL Targets One way people look at this is by focusing on their cost per conversion target. If you are willing to pay $10 per conversion, and you are paying that on some words but only paying $8 on others, you'd most likely be wiling to up your cost per conversion target to get more conversions, assuming that this increases your overall volume. With that in mind, let's take a look at the reasons how you could end up paying less per conversion than you are willing to pay: 1. You may have a problem due to capped bids. You could have bids that are ultra high performers that are prevented from being fully optimized due to a bid cap (that you
Pag e 2 . C o p y r i g h t © 2 0 0 8 M a r i n S o f t w a r e , I n c . applied). This can happen if you set too low a bid cap in your bid management tool, because you could be leaving traffic on the table. Bear in mind that the keywords that start approaching your cap are good performers by definition (assuming you are using a bid management tool, or if you apply some sort of cap in your Excel-based calculations). 2. You may well have some keywords that are already in the #1 or the #2 position. The good news is that means that the keyword is working for you, and you are getting a high percentage of the available traffic share. Note that being in the #1 position does not always provide the highest ROI, but it should almost always provide the most traffic. Related to this is the gap between the price you bid, and what you actually pay, sometimes known as "headroom."For example, you may bid $1.00 on a particular keyword but end up getting charged $0.93 per click ($0.01 over the next higher bidder, adjusted for quality). This gap averages about 7% and varies widely by industry and keyword. Again, these bothseem like a good thing (your keywords are profitable!) but the ironic consequence is that you might naively attempt to raise your bids without having any actual impact on position (or cost). 3. It also happens that the performance level changes vs the past. If you have holiday seasonality, you may have more buyers and fewer browsers than your data from the non-holiday season would indicate. Unless you make proactive adjustments, you might spend $10 for 10 clicks but it would yield 1.2 conversions instead of 1 conversion!Again this seems like a good thing but there may be even more traffic, again assuming rule #1does not come into play. As a result, you should have bid $12 in case there was more traffic available.There is a balance here between market responsiveness and building a statistically reliable data sample from your performance history. Pag e 3 . C o p y r i g h t © 2 0 0 8 M a r i n S o f t w a r e , I n c . Keyword Waterfall Analysis Let's take a look at the plot of some data from a real campaign for... [download for more]