By now everyone has heard the general consensus on common ad group setup practices. An ad group should have a limited number of keywords, be related to one another, and be relevant to the ads and landing pages. Each one of these items is important but leaves ambiguity in the actual implementation. If you spend any time thinking about these you’ll arrive at more questions than answers.
Today we’ll cover a few of the additional considerations for creating ad groups as well as the pros and cons for each. Some of these styles will be used alone but others could be layered or combined into something greater. The goal here is to loosen up the grip of generic common practices and consider what direction to move.
Match Type Focused Ad Groups
Match type segmentation is one of the most useful ad group styles and one that shows immediate benefits. Regardless of how you group your keywords, this ad group structure utilizes specific ad groups for each match type. For example, red shoes exists in both broad match and exact match versions in separate ad groups.
For this reason you’ll see tiered bidding strategies as match types perform differently. I like to keep them completely separate though and not just controlled by bids. Take a look at your keywords segmented by query match type and you’ll see that exact match queries don’t always trigger exact match keywords.
By segmenting match types, especially exact, you maintain tighter control over some of your most valuable keywords. Instead of relying solely on tiered bidding you can also exclude your exact match keywords from your broad match ad groups to funnel traffic to the specific ad group.
Although it won’t always be the case, you could see dramatic increases in performance by segmenting match types. You may be surprised by the boosts you see once you start treating the match types differently. Even with a basic segmentation keep an eye on your exact match impression share. It’ll often increase once you separate it out.
Top Performers and Query Farming
The next style is similar to match type segmentation but with a looser scope. Rather than segment all keywords by match types you only create focused ad groups for your consistently converting queries. This lessens the bloat in a campaign, by reducing the number of ad groups with little traffic, while giving the same benefits of segmenting.
To begin, you would start with broad or modified broad match keywords related to a product. Over time converting queries come in via the search query reports. Weekly or every other week you would then pull the converting keywords, exclude them from your general ad group and build a new more specific ad group for the converting query.
This style makes negative keyword management a breeze as negatives are localized to a few of the broad ad groups. This means less searching for poor performers across multiple ad groups, and minimizing those small local leaks that add up to a sizable chunk of spend.
Again, the benefits are similar to the match type segmentation but this style offers more flexibility. This is especially so for accounts that convert on an ever changing list of queries. By adding the convertors you’ll be prepared next time but you won’t spend so much time building ad groups that ultimately have little impact on the success of your account.
Organizing by Profitability
Beyond themes and match types, you may opt to sort your keywords by profitability. This is especially crucial if you develop a portfolio system for bid management. Rather than theme, keywords are sorted by one of the important conversion metrics such as the amount of revenue, profitability, or product margins. What you choose will depend on how you define PPC success for your account.
Organizing keywords in this manner allows you to step back and evaluate them as a group rather than individual keywords, which regularly fluctuate in performance. Organization is key here as you may opt for the default ad group level bid so you want to make sure that bid applies to each of the keywords.
One of the key benefits here is focusing on driving value at a higher level, rather than worrying too much about the specific details of each keyword. This means more time focused on performance, seeing the forest develop without the focus on any individual tree.
Even if you don’t plan to use a portfolio approach, this style is great for budget-capped accounts or those whose budgets change regularly. By segmenting your ad groups by the value they drive with each conversion, you make it easier on yourself to make cuts and pauses to lower value ad groups in the event you need to free up budget space for the more profitable groups.
Besides the organizational aspect, you can’t forget the other relevant factors you need to consider in any account style. The entire existence of an ad group is to show specific ads to a targeted audience. You should always keep this in mind while planning.
Relevant ads are not just for the searchers convenience. Google and Bing rate your ads on relevance to ad copy, landing pages, and extensions. Each of which are vital for top performance in the auctions. This is the main reason why ad group structures based on match types and tightly woven themes continue to stay in fashion year after year. Even going so far as single keyword ad groups.
One of the easiest methods for organization is fitting a great headline to the ad group’s ads. This is a great first step but shortsighted depending on your products. While the basic ad copy may line up with a group of keywords, do the sitelinks and callout extensions match as well? For instance, different types of siteslinks and specificity apply to different types of keywords. Of course it is difficult to get large samples in tightly focused groups but you can think through the general concept easily enough.
For example, you are advertising appliances and while there are many variations they ultimately perform the same with similar ads. If you take a step back you can see the missed opportunities. For example, general blenders should have sitelinks and callouts to more segmented pages to help define the options. While searches that infer a more refined interest, such as high capacity, should instead feature links to the different sizes of higher capacity models.
The same goes for callout extensions. If you are running a cloud services platform with a limited range of contracts, you can craft specific ad messaging pretty easily to highlight your top selling points but could even further refine your messaging with specific callouts regarding the service they are searching.
There is no one size fits all but the best rule of thumb is, “when someone sees this product or is looking to solve a particular problem, what else would I like them to know.” If there is any variation between the answers for two keywords, they should have different ads.
As with anything in paid search, the answer to these questions is always changing. With the plethora of competitors and services across the search engines, new styles will always appear. The best practices put you in the right ballpark but it is up to you to continue to develop your own system if you wish to reap the greatest rewards.
Feel free to leave a comment, question, or even a story on what worked or didn’t work for you. This topic is so personalized to individual accounts it’s always interesting to see other opinions.