Can the act of simply “re-organizing” an SEM campaign’s keywords have a positive impact on cost metrics? Can a campaign restructuring drive down cost per click (CPC) and cost per lead (CPL)? That’s the question I asked myself in early 2013, as I brainstormed new ways to improve cost efficiency while maintaining impact.

Based on A/B testing amongst our numerous SEM clients, the short answer is “yes.” Let me explain.

Zywave’s SEM (or pay-per-click) campaigns are comprised of “ad groups” that contain closely grouped keywords. If the campaign’s goal is to attract clicks and leads for “home insurance” customers, the SEM campaign focuses solely on home insurance keywords – thousands of terms that search engine users employ to get information on home insurance policies. Historically, that home insurance campaign has been further segmented into ad groups that organize a large group of keywords into smaller sub-categories. It was common to see ad groups for generalized ad groups that contained very broad keywords. An example of this would be an ad group called “house insurance,” which contained keywords such as house insurance, houseinsurance, insurance for house, insurance for houses. The ad groups became, I thought, unwieldy — there were too many keywords in each ad group to accurately monitor and identify the key lead drivers. It was not uncommon for each ad group to include 200 keywords.

Back to the A/B test. The hypothesis of the test was this: modifying the current campaign structure and keyword selection of our “Home Insurance” SEM campaign, will result in clients seeing improved cost efficiency (measured by cost per lead and cost per click).

The idea was simple: take the best performing home insurance keywords from our client’s campaigns, restructure their campaigns to include “tighter” (or, more closely grouped) ad groups.

An exhaustive keyword-level data analysis of historical campaigns ensued for about two months, the details of which I won’t bore you with, before the SEM team identified a slimmed-down keyword list of the top-performing “home insurance” terms.

The next step was to visualize and create a new campaign structure, one that grouped similar keywords into smaller, more manageable ad groups. I’ve always held the notion that the most effective keyword strategy focused on something called “user intent” – match your most important search terms to what it is that your audience is actually searching for. This was the crux of the new campaign structure; all ad groups focused on the intent of the keywords contained within. All terms were matched up with what our keyword analysis told us our most important users were looking for.

The new campaign structure is visualized below:

All keywords within the home insurance campaign (including all iterations of searches for home, house, homeowners, homeowner, home owner insurance) are segmented by what we believe the audience is trying to accomplish. Note the ad groups are segmented by words that indicate what is most important to the user: such as “buy,” “online,” “free quote.” All these terms give a higher level of targeting for our audience.

This restructuring allowed the SEM team to accomplish two important things. First, we were able to more closely report on specific and related searches to see which are driving cost efficiencies, and which are most costly. Secondly, since we are able to serve unique SEM ads to each individual ad group, our clients’ ads could more closely speak to the user’s intent. The “Online – Home Insurance” ad group would now feature ads that said “Looking for home insurance online? Come get a free quote for a new policy today.” Search users react favorably to SEM ads that match their search phrase.

So what happened?

The results were impressive. Over the course of 60 days, Zywave ran home insurance campaigns side-by-side and compared two important cost metrics (CPC, CPL). The simultaneous test included a group of 24 of our SEM clients, and resulted in data coming from 464,000 ad views. During that 60-day time frame, our tests showed that the “new” home insurance campaigns outperformed the “old” campaigns significantly:

• Cost per click decreased by 24.7%
• Cost per lead decreased by 29.8%

Two of our most important cost metrics showed significant improvements, allowing us to achieve one of the foremost tenants of the Zywave Internet Marketing department: to give our clients more impact for their money. In this case, cost efficiencies are allowing us to drive more leads with the same budget.

What’s next? You’ll have to wait for my next blog to see how Zywave implements these findings across all SEM clients.

Contact [email protected] with an inquiries or questions about new or existing SEM campaigns.