Today marks the third and final installment of our blog post series on PPC account segmentation. During the first post we looked at ways to segment your PPC account by content and location. Next, we looked at ways to Segment your ppc accounts by Device and Medium (Search or Display). Today, we will look at ways to segment your accounts by Network (Owned and Operated or Partner) and Match Type.
Network (Owned and Operated vs. Partner Network)
As we have previously mentioned, this feature is only available on Bing Ads. In fact, this is one area where Bing is beating Google. In Bing Ads you have the option to isolate and separate your search traffic between Microsoft and Yahoo Owned and Operated Sites and the Bing and Yahoo Search Partners. In addition, Bing offers advertisers the flexibility to exclude specific partner sites from the search network due to poor performance. They even give you the performance stats for each individual URL.
One of the first steps that I take when taking on a new Bing account is to separate Owned and Operated traffic from the Search Partners into separate campaigns. We do this by creating an exact replica of the existing campaign and naming the new campaign with the convention “Campaign Name_Partner”. You can then control the settings under the Ad Group Settings>Advanced Settings>Ad Distribution field. The existing campaign should be set to ‘Bing and Yahoo! search (owned and operated) only’ and the “_Partner” campaigns should be set to ‘Bing and Yahoo! syndicated search partners only’.
This set up allows you to manage the bids and budget for each traffic path separately based on performance. Often times, the bid required to hit a certain position or revenue target will be lower on the partner network than on Yahoo and Bing Traffic. If you are using an Automated Bid Tool, such as the one that we use here at Webtrends, this allows the system to have that extra level of granularity in hitting your revenue targets. If you are not using an automated tool, these changes can still be made manually. In addition, this set up will allow you to shift budget and focus to the channels that are bringing in the best return.
Segmenting your PPC accounts by match type may be a more advanced strategy than some of the others that we have explored in this series; but it is no less important. This is the real meat and potatoes of Search right here. It can also tend to be more controversial. One has only to bring the matter up on #ppcchat to see evidence of that.
Whenever you are setting up a new campaign, or building a new account from scratch, you will have to make a decision about how to handle Match Types. The most basic of strategies, and the one that we would least recommend, would be to slap the most broadly desired match type on a keyword and forget it. I shudder any time I review an account and find hoards of keywords with bids on Broad Match only. When you decide to bid on broad only, the bids for the lower match types (phrase and exact) become inherited from the broad. The data metrics that you are reviewing are really an aggregate of all 3 match types. Accordingly, you have to optimize and bid towards that aggregate. When you adopt the strategy to bid only on the highest match type you lose the ability to optimize based on differing performances of match types. For example, I often find it advantageous to bid higher on Exact Match than Phrase or Broad. This makes sense as Exact Match Mappings should naturally be more relevant and the major Search Engines tend to give Exact Mappings preference in Ad Rankings. By bidding on only one match type, you lose that sense of control and granularity. Nobody wants to be that guy.
A popular strategy used among many PPC experts is to separate your match types into separate campaigns or ad groups. Either way, the strategy remains the same. Let’s say that you have the keyword ‘Free Cheese Pizza’ located in your ‘Free Pizza’ ad group and want to bid on Broad, Phrase and exact. With this implementation strategy you would have 3 separate ad groups (or campaigns) for each Match type. Here you would have ‘Free Pizza Broad’, ‘Free Pizza Phrase’ and ‘Free Pizza Exact’ Ad Groups. In the ‘Free Pizza Broad’ ad group you would place a bid on Broad for your ‘Free Cheese Pizza’ keyword. A phrase bid would go in to the phrase ad group and so on and so forth.
Going this route does give you the greatest control over your match types. This methodology makes it the easiest to review data and optimize by match type. You even have the option to write different ad copy by match type. However, if you are going to go this route, it is absolutely critical that you use negatives to direct traffic. For me, that is where the rub comes in. In order to ensure that only broad traffic goes into your Broad groups, phrase traffic in your phrase groups and exact traffic in your exact groups you have to be very meticulous about your negative keyword implementation.
As previously mentioned, when you bid on broad match you are also implicitly bidding on phrase and exact match. This means that your ‘Broad’ ad group also has numerous bids on phrase and exact even if you do not explicitly state so. Accordingly, you could serve phrase or exact mappings in your broad group and cause additional competition for your keywords inside your own account. In order to prevent this, every keyword in your broad campaign will have to be added as a phrase negative in that group. Similarly, every keyword in your phrase group will need to be added as an exact negative in that group. This prevents phrase or exact traffic from flowing in to the broad group and exact traffic from flowing in to phrase. For more on Negative Keyword Management and Match Types, see John O’Toole’s latest blog post.
While this strategy does provide the greatest level of control, something that I am generally a large advocate for, one can see where it can also quickly become burdensome. Managing the negative keyword structure in accounts with thousands upon thousands of ad groups can become a chore. Nowhere is this more true than in the realm of keyword expansion. Any good PPC account manager will routinely review Search Query reports for new queries that can be added as keywords. Under this methodology, adding a keyword suddenly becomes a multi-step process involving 3 separate ad groups, 3 separate keywords and two separate negative keywords. Scaled out, the process tends to become burdensome.
Because of this, I tend not to recommend segmenting your match types by group or campaign. The benefits are great. But the cost in time is often not worth those benefits. Ultimately, it comes down to the particulars of your account and your current workload. If you can manage to do this segmentation and still do a healthy level of what I would consider more important PPC optimization pieces – Keyword expansion, ad testing, bidding, etc go for it. If not, best look for an alternative.
My preferred method of breaking out match types is to have an explicit bid on all 3 match types for any individual keyword contained in the same ad group. To go with our example above, you would have a single ‘Free Pizza’ ad group. In that ad group would be 3 different bids for the keyword ‘Free Cheese Pizza’ – One each for Broad, Phrase and Exact. This gives you the freedom to bid and optimize by match type without overburdening your time with negative keyword management. When starting out with this strategy I always take a 100/85/75 methodology for bidding. That is, the bid on exact match is the starting point with the phrase bid 85% of exact and the broad 75% of exact. From there, you can optimize your bids based on performance. As a side note, it should be noted that when I refer to ‘broad match’ I am really suggesting Modified Broad Match. But that is a matter for another blog post.
In the end, the management of your match types really boils down to the particulars of your account, the goals you have set up and the available time you have for management. If you have the time, by all means segment by ad group. For others, that may not make as much sense. At the very least, being cognizant of Match Types and how you will segment them is critical in the early planning phases for any new campaign or account launch.
In what ways do you love to segment your PPC Campaigns? Are there any that we may have missed? We would love to hear your feedback and suggestions.
For more information on how Webtrends can help you segment and manage your PPC Campaigns, please visit our website or contact Chase.Wells@webtrends.com.