Below is a review of key metrics provided for Naytev tests. Primary results for tests are available in the 'Test Results' view. Much more detailed results are available using Naytev's test-level excel export (available in 'Test Results' view), and team-level excel export (available in the 'Tests' menu).
All Test Types
Score: Naytev score (0 to 100) is a calculation of relative message performance based on net results across key metrics for a given optimization goal. This score is available for all messages with at least 250 impressions. The weight of each metric is dynamic, with more weight assigned to metrics where there is greater performance differences between messages. For example, CTR results will be automatically given more weight if there is a large performance difference across messages, such as a CTR range from 1 - 6%. If the range is smaller, such as 5 - 6%, CTR results will be given less weight. This ensures weighting is fair and clearly identifies which message is the best net performer. The Naytev score is adjusted based on the test goal, such as Post Engagement and Video Views.
Estimated Relevance: this metric is accounted for in the Naytev score, and reflects how well Facebook estimates your message is resonating with your target audience. Messages are characterized as having a higher estimated relevance when they have a high frequency of positive, rich engagement. Facebook serves the most impressions to messages with the highest estimated relevance.
Impressions: Facebook serves more impressions for better performing messages, and phases out worse performing messages. It is normal for some messages to receive significantly more impressions than other messages with lower performance. Some messages with limited data may report high performance, but may still be phased out if their estimated performance is much lower than that of other messages.
Engagement: includes Page Likes, Post Reactions (Like, Wow, Haha, etc.), Post Comments, Post Shares, Post Clicks, and Post Video Views. You can see the breakdown of engagement by hovering over engagement values in your test result page and in the Naytev data export.
Engagement Rate: equals the sum of all engagement divided by total impressions.
Link Tests Only
Click-Through Rate (CTR): equals total link clicks divided by total impressions.
Cost Per Click (CPC): this is a very important indicator of a given message's breadth of engagement. A lower CPC translates to better performance across your target audience. For example, assuming two link post messages share a comparable CTR, the message with a less expensive CPC will drive more referral traffic than the other message when shared. The higher cost message may have good CTR during the test, but its higher cost indicates it may only maintain that rate with a smaller subset of your target audience. The other message's lower cost indicates it will better maintain its CTR across a much larger portion of your target audience.
Video Tests Only
10 Second Views: the number of times your video was viewed to 10 seconds. If your video is less than 10 seconds than this refers to the number of times people viewed 97% of your video. This metric is more robust than 3 second views, since a large volume of 3 second views result from videos auto-playing. 10 second views better indicate how many people are very interested in your video. Naytev provides 3 Second Views in our data export.
10 Second View Rate: equals 10 second views divided by total impressions. This is the benchmark view rate performance metric. Naytev provides 3 Second View Rate in our data export.
Cost Per 10 Second View (CPV): this is a very important indicator of a given message's breadth of engagement. A lower cost per view translates to better performance across your target audience. For example, assuming two video post messages share a comparable view rate, the message with a less expensive cost per view will drive more views than the other message when shared. The higher cost message may have good view rate during the test, but its higher cost indicates it may only maintain that rate with a smaller subset of your target audience. The other message's lower cost indicates it will better maintain its view rate across a much larger portion of your target audience.
Average Percent Watched: the average duration of the video watched divided by the total video length. Naytev provides a breakdown of average seconds viewed and video retention over time in our data export.
Image/Text Tests Only
Cost Per Engagement (CPE): Total Spend divided by Engagement Activities. A low cost of engagement indicates that you're effectively engaging with a broad portion of your audience.