The Overview section at the audience level is the first thing you will see when you load a report. This section provides a glance at your entire audience by summarizing the key insights found in each cluster’s Interest, Member, and Content sections. As with all of our sections, the data varies from network to network. In this article, we're going to cover the main data points you'll find in the Overview section of each report.
Report Name
Regardless of network, you will always have the option to change the name of a report. Simply type over the existing name and hit the "Change Name" button.
Audience Details
Here you can see the date range covered by the report, audience size, average number of posts, and average interest similarity. Interest Relevance is the average Relevance score for an audience's top 100 interests. This number gives you an idea of the homogeneity of an audience's interests (i.e. if an audience shares many common interests).
Audience Query Source
Here you will see which parameters were used to filter the audience captured within the report. This is particularly useful if you are viewing a report that someone has shared with you.
Audience Visualizations
Network Graph: for reports with unsupervised segmentation, this diagram represents how the different clusters come together as an audience. The lines connecting clusters indicate their shared interests; the more lines, the more interests they have in common. Any white space between clusters indicates that there is less of a connection.
Cluster Sizes: a pie chart that shows the size of each cluster within the total audience. Hovering over a slice of the pie will show you the number of people within each cluster. This visualization is available for all of our network options.
Audience Summary
Lurker Percentage: this shows the percentage of an audience who actively post as compared to those who lurk (i.e. seldom post, or only consume content). We define lurkers as those who post fewer than 15 times per month. This graph is available for Twitter reports.
Posts (or Tweets): this shows the average number of posts each cluster member makes per month. For Twitter reports, we display the 25th, 50th, and 75th percentiles to give you an idea of what normal activity looks like for this type of Affinio report. These percentiles are based on data from Affinio's repository of reports.
Interest Relevance: this metric shows the likelihood of shared interests within a cluster on a scale of 0-100, where 15 indicates a cluster of like-minded people, and above 40 is exceptionally high. This metric will help you determine a cluster’s quality or homogeneity.
Unique Interests: this shows the proportion of interests in a given cluster that are completely unique to that specific cluster and not shared by the rest of the audience.
Interconnectivity: this metric shows how friendly or well-acquainted a cluster is. The higher the density percentage, the greater the likelihood that people in this cluster know (and follow) each other. When publishing content, keep in mind that a higher density indicates cluster members are seeing the same content shared from multiple people. This graph is available for Twitter reports.
Share of Images: this shows the percentage of images shared by each cluster relative to the total number of images shared by the audience. It does not represent the percentage of tweets that included images. This data comes from Image Analysis, which is found in the Content section of Twitter reports.
Cluster Summaries
See condensed summaries for each cluster that include the top three distinguishing traits and a preview of unique interests. The traits are pulled and scored from across member bios, interests, locations, and content, and are qualified based on cluster depth and relative audience use. The mini interest mosaics offer a preview of the interests that are unique to a particular cluster, and are not shared by the rest of the audience.
You can use the Filter trait saturation (traits level) in Cluster Summaries, to see how the top results change when you require more minimum member coverage. The saturation here represents the minimum cluster (versus audience) penetration, since trait cards are inherently at the cluster level.
Baselines
The Baseline feature applies an audience-to-audience comparison for context. Two different lenses are provided to nuance and score the report you are viewing against another report: (1) a general audience-to-baseline assessment of distinctiveness (labelled ‘Audience Comparison’), and (2) a more specific cluster-to-baseline assessment of distinctiveness (reflected on each Summary Card).
Read more in the Baseline FAQ.