Google made 116 billion USD in revenue from ads alone in 2018. Advertisers spend a considerable amount of budget on display advertisers, yet spend less time on reporting and optimizing. In this list, we will cover all the primary metrics you need to know to analyze your display campaign. Let us get started:
Disclaimer: The metrics are not listed on the basis of importance.
The impression is when the advertiser’s ad loads on the user’s web browser. Impression counted on the server does not imply that the user has viewed the ad. And when the user reloads the browser, if the same ad appears, then it is counted as a second impression. It simply means that the ad was rendered by the user’s browser.
The number of impressions the ad is eligible to receive based on the chosen targeting and bidding strategy. In simple words, it shows the Ad potential. The narrower the targeting is, the smaller the eligible impressions you can get. However, this is not the only factor influencing the metric, quality of the ad and the bid plays a major role.
The number of impressions that were viewable by the user is called viewable impressions. If at least 50% of the ad was viewed for one second, then it is classified as a viewable impression. Viewable impressions help the advertiser to analyze the quality of the impressions.
As you might have guessed, it is the opposite of viewable impressions. The display ads wherein even 50% of the ad was not viewed for a second, it is classified as a nonviewable impression.
Measurable impressions are the number of impressions that could be measured by Active view technology. Active view technology is used to measure if the ads are viewable by the users and are widely used by Youtube and other display networks.
Nonmeasurable impressions are the number of impressions that could not be measured by active view technology. However, this does not imply that the display Ad was not viewed by the user.
Impression share is calculated as Impressions/Eligible Impressions. This is a competitive metric which shows how much % of impressions you could have got if you had more budget. It gives you a roadmap for your bidding strategy for campaign optimization.
Lost Impression Share
This is the % share of the impressions that were lost due to insufficient budget. This metric gives an overview of lost impressions to your competition. Same as above, good for reviewing the bidding strategy.
Viewable Impression Distribution
Viewable impression distribution is a percentage of viewable impressions to the total number of impressions. It is calculated using the following formula:
Viewable Impression Distribution = (Number of Viewable Impressions)/(Number of viewable impressions+Number of measurable Impressions)*100
It is important to note that this should not be confused with viewability.
NonViewable Impression Distribution
NonViewable impression distribution is a percentage of Nonviewable impressions to the total number of impressions. It is calculated using the following formula:
NonViewable Impression Distribution = (Number of NonViewable Impressions)/(Number of viewable impressions+Number of measurable Impressions)*100
Measurable Impression Distribution
Measurable impression distribution is a percentage of measurable impressions by active view technology to the total number of impressions. It is calculated using the following formula:
Measurable Impression Distribution = (Number of measurable Impressions)/(Number of measurable impressions+Number of nonmeasurable Impressions)*100
Nonmeasurable Impressions Distribution
Similar to the above, nonmeasurable distribution is evaluated as a percentage of nonmeasurable impressions to the total number of impressions. It is calculated as follows:
Nonmeasurable Distribution = (Number of measurable Impressions)/(Number of measurable impressions+Number of nonmeasurable Impressions)*100
This is one of the most important metrics and is a primary display KPI. Viewable rate as implied by the name indicates viewability of the display ads. It is calculated using the following formula:
Viewable Rate = Number of viewable impressions/ Number of measurable impressions.
This gives an idea of the % of ads that were measurable by active view technology. The measurable rate can be calculated as a % ratio of the measurable impressions to the total number of impressions served by the ad server.
Unique reach is the number of users who interacted or viewed the display ad. This measures cross-device information using Google sign in thus providing accurate info as compared to the cookie reach. Keep in mind that it is not 100% accurate as sometimes, cross-device info cannot be attributed or traced back to a particular user.
The Ad servers use a cookie to track user interactions. The cookies are specific to the device and can be attributed to the user as long as the user uses one device. Cookies are specific to desktops as interactions on the mobile web are accomplished using the device IDs. The cookie reach gives the unique reach of impressions per cookie. It may not be accurate to the attribute cookie to a user anymore as almost every internet users use more than one device(mobile, laptop, tablet, console, connected TV, etc)
Frequency is defined as the number of times one Ad was viewed by a single user. It can be calculated as Impressions/Reach. This data is available as “Average Frequency per Cookie” in the reporting tool. It is recommended to set the frequency between 3-5 as this is a perfect balance for better brand recall and annoying the users by showing the same ad.
Average Viewable Time
Average viewable time is measured as the time in seconds the viewable ad was viewed by the user. The active view tech stops counting the time after 3 minutes for display ads.
As simple as it sounds, clicks are registered when the user clicks on the ad. Though clicks are a good measure of user’s engagement with the ad, it does not provide a full picture. Some % of total clicks are accidental due to the small screen(especially on mobile). Hence it is always good to measure the quality of clicks by directly comparing it with the number of sessions coming from the campaign. (P.S do not forget to add UTM parameters to your click-through URL on the display Ad, more optimization tips here)
Google uses algorithms to classify the click data into valid and invalid clicks. The clicks Google identified as spammy malicious clicks from bots and virus are grouped under invalid clicks metrics. This metric also includes the second click of the double click on the ad by the user. Based on the buying model, this significantly decreases the advertising costs and keeps the account non-spammy and optimized.
CTR abbreviated as Click-through-rate is calculated using the following formula: CTR = (Clicks/Impressions)*100. It is always expressed in terms of a percentage. This is a quality KPI which provides a vague picture of the engagement rate of the users with the ad. It signifies the probability of the ad getting clicked. A lot of factors influence the CTR of the display banners right from the targeting strategy(Ad relevance) to the creative design (Attractiveness/compelling message) of the ad itself.
This provides quality info as compared to CTR as it takes viewable impressions as the base. As 100% of the impressions are not viewable, it gives a fair picture of the true interaction rate of the user with the ad.
The action that the user takes that is profitable for your business after the user interacts with the ad(views, clicks, watches the video, etc) is called a conversion. A conversion is defined by the advertiser and could be anything from downloading an ebook to making a purchase. It is very important to set up a campaign with a conversion goal in mind. Following this, do not forget to set up conversion tracking. If you want to know how it is done, you can find it here.
Click through Conversions
If the said conversion happens within 30 days after the user has clicked your ad then the conversion is called click-through conversion. The 30 days is a default lookback window, you can custom set it up to 90 days.
View through conversions
As the name suggests, if the conversion happens within 30 days after the user has viewed your ads then the conversion will be assigned as a view-through conversion. Same as above the lookback window can be set up to 90 days depending on your business/user history and interactions.
It is the rate of the ratio of conversions by interactions. For example, if you would like to the conversion rate of the campaign wherein the conversions were set to “purchase”, then the conversion rate would be calculated as follows:
CVR = (Conversions/Clicks)*100.
ROAS abbreviated as Return on Ad Spend gives a picture of how the campaign is performing on a holistic level. It is calculated using a simple ratio: Revenue generated by the campaign/Cost of the campaign. For example: If you spend 200 USD on the campaign and you generate 2000 USD in revenue(not profit, just revenue), then your ROAS will be 10:1 or 1000%
Revenue per Click (RPC)
As the name suggests, revenue per click is defined as the total revenue generated from one click. It is calculated as follows:
Revenue per Click = Total Revenue generated from the campaign/Total number of clicks from the campaign.
Revenue per Thousand Impressions (RPM)
Revenue per thousand impressions if the revenue generated for every thousand impressions of your display ad. You can calculate Revenue per thousand impressions using the following formula:
Revenue per mile = (Total Revenue generated from the campaign/impressions)*1000
Click through Revenue
The total amount of revenue generated from click-through conversions can be classified as a click-through revenue. This metric helps you evaluate the worth of the click and needs to be monitored and optimized over time.
ROI(Return on Investment)
A primary metric for the marketers as we need to make sure that we are not losing money from our campaign advertising budget. In order to calculate the return on investment, you need to know all the incurring costs as well. You can calculate ROI using the following formula:
ROI = (Revenue generated from the campaign – Cost Incurred) /Cost incurred
or ROI = (Net profit)/Total Investment
Abbreviated as Cost per click, it is one of the buying models of display ads which is also called as Pay per click. It is the cost you are willing to pay as an advertiser for one click. CPC is a bidding strategy as well wherein you could manually set a CPC to fixed cost or automated CPC which will maximize clicks until your budget is exhausted
Cost per mile, or Cost per thousand impressions is a buying model where you pay a fixed/dynamic price for a thousand impressions. CPM is generally a buying model for brand awareness campaigns or to promote an event. CPM can be calculated using the following formula:
This has a different meaning to the publisher and advertiser. It is simple from the publisher’s side, it just means how earnings per thousand impressions i.e, how much money did the publisher make for displaying the ad 1000 times. From the advertiser perspective, it is used as a linear model to directly compare the CPMs across channels and is calculated as follows:(Media cost *1000)/Impressions
Average Viewable CPM (vCPM)
At this point in time, you already know that not all impressions are viewable. This is a bid strategy widely used by Adsense to bid on placements that have high potential to become viewable. So you would only pay for those impressions that were deemed to be viewable, i.e, 50% visible for at least one second. This is a preferred bidding strategy for a brand awareness campaign wherein the goal is to maximize the display ad views.
CPA (Cost per Action)
Cost per action is an eCommerce centric bidding strategy. You could set a conversion goal and only pay when the desired “action” is completed by the user. The average cost per action can be simply calculated as the total cost of the campaign/total number of conversions.
This is a bidding strategy to maximize conversions within the set target CPA. This does not mean you pay the same CPA for each conversion. Some may be higher, and some lower, in the end, it will balance it out to reach the set target CPA. However, for this strategy to work, the conversion tracking needs to be set up.
Above the fold? Below the fold?
This is an age-old term which was basically used by the print media. The section that is visible to the users without scrolling is “above the fold”. The section that is visible to the user after scrolling is called below the fold. Using Google Ads exclusions, you could exclude all the placements that appear below the fold. However, test this for a couple of days and confirm if below the fold placements do not work for you before making a final decision on exclusions.
This is a good insight to add to your client reporting on which OS worked the best for your campaigns. For example, we have observed that for premium brands, maximum impressions/clicks were from iOS. This is not really important and is a dimension just like platform as opposed to the other ones listed here which is a dimension.
In the reporting section, you will be able to view the device split of impressions served (and other metrics too). While running the report template, make sure to add “platform” as a dimension to get the device split of the available metrics. This helps in preparing the strategy for future campaigns. For example: If 80% of your impressions are from a mobile platform, it would be a good idea to create a mobile-only campaign.
Now you have mastered all the important terms and metrics that are important to analyze the display campaign performance. It is a good idea to bookmark this page as a reference and read it from time to time to refresh your basics. Do not forget to check out best practice for display ad creatives to optimize your next display campaign performance.