In almost 15 years of honored PPC management service there has always been a key question to answer, especially when checking performance of a long run account.
How much optimized is this? Or, in other words: is there any margin to improve results? And, immediately after: where should I began working?
After having answered hundreds of times to the same questions I developed a methodology to audit Google Ads campaigns and finally, in 2016, I patented it). Here is how “PPC CheckMate” methodology works.
This quick and reliable methodology will answer these questions in minutes on both simple or very complex campaigns (regardless their structure and networks used) looking at:
- Efficiency – Are we buying clicks at the lowest possible cost?
- Effectiveness – Are we buying the best clicks (the ones that “convert”)?
- Coverage – Are we getting all the best clicks we can?
Efficiency is the easiest and fastest KPI to verify (at least in the Search Network). Just activate the Quality Score custom column and order your keywords (at account, campaign or group level) by descending total cost (in a time lag large enough to have consistent data).
If the majority of your top spending keywords do have a QS of 8 to 10, you have a structured your account well enough (in Google’s eyes). The lower the scores are, the more optimization work you will need.
On the Display Network you can only indirectly deduce QS looking at Click Through Rates (the higher, the better -exception made for low quality or unrelated sites/placements), Cost per Click (which should be significantly lower than in Search Network, except for highly focused placements) and Bounce Rates (which can be imported from your Analytics account).
Before auditing for effectiveness, you will need to verify that all possible conversion actions are tracked in your landing site (and imported in Google Ads, if tracked in Analytics).
In the long run no effective PPC account can afford to spend in an ad-group more than you gain from it, unless you do not want to donate part of your marketing budget to Google.
There are some caveat to that. The first is that is not always easy to set a correct value for leads. In branding sites you should always consider behavioral conversions (time on site or PDF downloads or video views, etc) and even when you have online sales, you should keep into consideration the average lifetime value of a new customer, not only the first “shot”.
Second. Technical attribution problems will always depress real conversion figures, unless you use the native Google Ads tracking code.
Third. Do not underestimate “discovery effect” and “word of mouth” value of apparently not-converting ad-groups. Generic search queries very often have an indirect influence on branded searches. If you stop them and wait for a wile you will be able to evaluate it.
Fourth. Always consider statistical relevance of data (the effectiveness of a keyword/placement should not be assessed without recording at least thousands of impressions)
Due to all these factors, instead of focusing on a strict assigned goal and stop whatever goes beyond it, you should investigate on “out of scale” ad-groups, indulging as long as it is economically sustainable on searches and display placements that are important for your branding purposes.
Last but not least, coverage in terms of impression share is a key factor in deciding where to spend your money.
Made 100 all the potential impressions you may have with unlimited budget, Impression Share simply tells you what part of them you are actually covering.
You should always activate these custom columns (available both for Search & Display Networks) and closely monitor them. All lower cost per conversion ad-groups (your digital “cash cows”) should always have the highest possible coverage. If this does not happen, your best ads are simply not being shown to anyone.
This may happen for several reasons, but the most frequent are: a) budget is too low for all the active keywords or b) CPCs are not set properly (low converting items are draining the budget which should go to the best ones).
Performance superstars should always have enough budget. A good equilibrium between daily budget and real expense in these campaigns is the only way to ensure people really interested will see your ads where and when they proved to be more successful. Search peaks for these ads are your performance’s best friends. Coverage leaks, the worst enemies.
Here is a quick recap:
Maximize Quality Scores of your top spending items creating ad variants, using all reasonable extensions, selecting the best keywords (and pausing too generic ones), using negative keywords (search queries report is your best friend), testing different match types, differentiating ads for mobile users, using local qualifiers, etc.
Look out for low converting items. To better performance consider tracking behavioral conversions, use conversion related automations (enhanced CPC, CPA, Dynamic Search Ads, etc.), checking CR on different devices (and modify bids accordingly), running A/B tests on your landing site, testing remarketing (both in search and display), blocking ineffective placements or click robots, etc.
Increase coverage of the best converting items, tuning CPCs to balance daily budget with expense, adding budget on your digital cash cows, regrouping campaigns to have in them more homogeneous items, scheduling extensions and ads visibility, etc.
You will find an infographic recapping this methodology at https://www.adworldexperience.it/en/measure-optimization-adwords-account-infographic/
Let me know what are your thoughts about this methodology, and, May the ROI be with you! 😉