Part 1 of this introductory series covered spending intelligently as you begin your journey in PPC marketing. We took you step-by-step through how a user on Google would find his or her way to your website via paid search ads. We also highlighted the importance of analyzing each individual step to minimize unqualified users from costing you clicks.
On to Step 2: Then you get the power
Be empowered by the data acquired in Step 1 and scale your online marketing strategy. In Part 2, we will delve into testing concepts from Part 1 to further develop and expand your PPC strategy.
Our friend, Tony Montana, didn’t start out as a powerful man, but leveraged his resourceful and scrappy ways to elevate himself to a position of power. Tony’s knowledge of his product, customers and competitors helped him stay in power.
This section will be the most helpful after you’ve spent some time and money on a PPC test buy and will assume you have data readily available for analysis.
We will start at the top of the conversion funnel and work our way down: Impressions -> Clicks -> Conversion
I hope you are ready to roll up your sleeves and spend a little time analyzing results from your PPC test, specifically looking at KPI’s (key performance indicator) from your campaign.
Tip #1: Try to avoid “Analysis Paralysis” by over analyzing every single metric. Create a story that can be explained using your KPIs instead.
A. Impressions vs. Impression Share (IS):
Impression: when an ad is shown to a user.
Basic question: How many impressions did my campaign generate?
Looking at the count of Impressions will provide you the total number of instances where keywords triggered ads to be shown on a search engine results page (SERP). When a search is performed in Google for the phrase “Hawaiian vacation with kids”, the following advertisers (1-3) each increased their impression count by one (+1) because of the search.
While the gross number of impressions your campaign receives may seem important, it is much more useful when you have additional data points that provide context.
Impression Share: the percentage of times your ads were shown (i.e. your total impressions) out of the total number of page impressions (i.e. pages where your ad appeared or could have appeared) in the market you were targeting.
Advanced question: Did my campaign get enough impressions?
This question looks beyond the raw number of impressions and addresses the more important goal of making each impression matter.
- Were your ads shown in a position that met your expectations?
- Most importantly, did your ads serve when you expected them to and how does this compare to your competitors?
Generally speaking, “ad position” is influenced by the amount you are willing to pay (max CPC bid) and the relevancy of the ad to the keywords in your ad group (Quality Score). Quality Score is a numeric representation of the relevancy of your ads and keywords assigned independently by both Google and Bing. It is important to note that only Google’s Quality Score impacts ad position currently. Bing’s Quality Score serves only as a guideline to improve your ad/keyword relevancy. We will discuss Quality Score in further detail in Part B.
To answer the second question, turn your attention to the Impression Share (IS) metric in the Adwords interface.
Tip #2: Impression Share stats can only be generated at the Campaign level (in Adwords) and is not available yet in AdCenter (as of May 8, 2011). Keep this in mind as you structure your Campaigns and Ad Groups. Impression Share metrics will be less useful in Campaigns consisting of Ad Groups with little to no relevance to one another.
Tip #3: To enable Impression Share metrics in your Adwords interface, 1) Select a Campaign, 2) Click on the Dimensions tab, and 3) Check all boxes.
In plain English, the above can be translated as:
- Impression Share: My ads showed up 84.41% of the time they were qualified to show on Google
- Exact match IS: If all of my keywords were set to “Exact Match”, my ads would have shown up 9 out of 10 searches (89.95%) for keywords in my campaign
- Lost IS (budget): My ads did not stop serving during this time due to my budget maxing out
- Lost IS (rank): Ad Rank = CPC bid x Quality Score. My ads did not serve 15.59% of the time because either my CPC bids or Quality Scores were low
As illustrated in the example above:
Impression share + Lost IS (budget) + Lost IS (rank) should always = 100%.
If you want to reach your potential customers at all times, your goal should be to have an Impression Share number close to 100% and the two “Lost IS” numbers close to 0%.
If your Impression Share is far from 100%, here are some tips to help you recover your “lost” impressions:
- Lost IS (rank), aka “CPC bids too low” – Look at the average position of each keyword. Is it above minimum bid (AdCenter calls it minimum bid, Adwords calls it “first page bid”)? With bids below these thresholds, your ads will not show up on search results pages.
- Lost IS (rank), aka “Quality Score too low” – Are your Keyword Quality Scores (QS) below 5/10? Per Google, since “Ad Rank” is a calculation of your Bid and QS, it would behoove you to improve Quality Scores by focusing on increasing CTRs in your Ads and Keywords. Improve CTRs by tightening up Ad Groups that only consist of closely related keywords and Ads that are the most relevant to these keywords.
- Lost IS (budget), aka “budget too low” – Do your campaigns have set daily/monthly budget caps? If so, are your campaigns hitting their caps frequently? Budget caps help pace PPC spend but can also suppress yours Ads from being shown if set too low. Google calls it “throttling” where Adwords won’t serve up ads every time they are eligible to be shown in an effort to allow your account to evenly pace through the daily budget.
How does this Impression data help empower you?
Now you know the difference between impressions and Impression Share (IS). Regularly monitor your Impression Share metrics and quickly fix issues as they arise. Low Impression Share hurts your chances at success by allowing your competitors to gain greater market share. Chances are, your competitors are already closely monitoring their IS and actively optimizing to 100% Impression Share. PPC is a dynamic platform – always look for opportunities to make gains over your competitors.
B. Clicks: CTR and QS (Quality Score)
Clickthrough Rate: number of clicks your ad receives divided by the number of impressions.
Basic Question: What is my average Click-Through Rate (CTR)?
At a high level, Clickthrough Rate gauges how convincing your ad is to your target audience. Many marketers often focus only on writing convincing ads. An equally important aspect to CTR is to target the ads to the right audience.
Steps to build a PPC campaign include:
- Creating Ad Groups
- Selecting and grouping keywords into Ad Groups
- Creating Ads for Ad Groups.
As discussed in Part 1 of this series, every Ad Group should consist of closely related keywords and a minimum of two ad variations. And as suggested in Tip #2 , every campaign should be comprised of closely themed Ad Groups.
This all seems pretty elementary… what’s my point?
Tight keyword groups (target your audience) + relevant ads (be convincing) = higher CTR
Don’t just build – build your campaigns and ad groups intelligently.
Take a look at the following sample Ad Group:
Given that the “theme” of this Ad Group is “Hawaii Vacation”, the keywords in this Ad Group seem pretty relevant to each other and the two Ad variations also appear to be pertinent to the keywords. Is this a targeted Ad Group?
Now look at Example #2:
If you are searching for the term “flights to Hawaii”, which Keyword-Ad combination is most likely to win your click? If you chose Ad Group 2, you are correct for the following reasons:
Assuming keyword bids are all equal, why is Ad Group 2 most likely to outperform Ad Group 1?
- Tight keyword groups – all keywords in Ad Group 2 pertain specifically to airfare
- Relevant Ads – both versions of the Ad pertain to airfare
- Display URL is more relevant to airfare
- Higher CTR will result in higher Quality Score, which, in turn, improves Ad Rank and lowers CPCs, which enables you to bid to a higher position**. Come again?
**Note: the following is a simplification of the relationship between CTR and Google’s Quality Score and should only be used as a high-level reference. Please refer to Google’s Quality Score documentation for additional detail.
Advanced Question: Why does CTR matter?
We discussed earlier that Clickthrough Rate is a measure of how convincing your ad is to your target audience. Google calls this “relevancy” and assigns a numeric value called the “Quality Score”
Per Google, Quality Score is made up of the following the core components:
- The historical clickthrough rate (CTR) of the keyword and the matched ad on Google; note that CTR on the Google Network only ever impacts Quality Score on the Google Network — not on Google
- Your account history, which is measured by the CTR of all the ads and keywords in your account
- The historical CTR of the display URLs in the ad group
- The quality of your landing page
- The relevance of the keyword to the ads in its ad group
- The relevance of the keyword and the matched ad to the search query
- Your account’s performance in the geographical region where the ad will be shown
- Other relevance factors
It is no accident that CTR is mentioned in three of the seven core components listed above.
CTR matters because it is a metric that can be controlled by marketers. However, while Google’s emphasis on CTR should be noted, it is also important that marketers don’t get tunnel vision with improving CTR. It is not an uncommon mistake for marketers to focus primarily on improving CTR… to their detriment. Creating highly attractive ads for the sole purpose of increasing CTR could be a costly error that ultimately impact your account history, especially if the ads are misleading and result in high bounce rates.
Note: Bing recently introduced their version of Quality Score. As stated on Bing’s site, Quality Score “is not used to calculate ad position directly” but indicates the degree to which ads are “eligible” to be served. At this time, it appears that only ad with a “Poor” Quality Score may have problems serving. Ads with “Good” and “No Problem” scores are eligible to serve.
The main takeaway is to define your audience with you selection of keywords and show ads that are relevant and compelling to that audience.
Conclusion: Get the Power, Use the Power
To some marketers, Impressions and Clicks are just simple one-dimensional metrics. You are not that marketer. Like Tony Montana, you know how to target your customers and are cognizant of your performance vs. competitors (Impressions Share),
At this point, you should be familiar with the following PPC tactics:
- Selecting and targeting your audience with tight keyword clusters
- Structuring tight campaign ad groups
- Building solid ads that are highly relevant and compelling
- Analyzing impressions and impression share
- Optimizing Ad Groups for high CTR, Quality Score and Ad Rank
I challenge you to apply these learnings to your campaigns and try to find the setup best suited for your conversion goals.
- Test – Test your campaign & ad group structures; test your ad copy variations
- Learn – Familiarize yourself with Impression Share, Quality Score and Ad Rank; use these metrics as a feedback mechanism to improve your key metrics
- Optimize/Refine – Modify what doesn’t work, test new ideas; repeat steps 1 through 3
Stay tune for the final part of this series that will dive deep into conversions, conversion metrics and optimization tips.
How KISSmetrics Can Help
About The Author: Jacob Shin is the Director of Online Marketing & Customer Acquisition at Savings.com where he focuses on high volume paid search strategy and conversion optimization for US and UK based traffic. Connect with Jacob.