Station Clusters

  Previous topic Next topic JavaScript is required for the print function Mail us feedback on this topic! Mail us feedback on this topic!  


The use of Station Clusters allows you to aggregate song score data from two or more databases, and create a compiled report from the combined stations.


Here's how it works:


First, decide which database you will designate as your main database from which to access Station Cluster reports. This can be any database on any server, but as the Cluster and Profile information is stored in a specific database, all Clusters should probably be created and maintained in the same database directory on the primary Fileserver.  (The database where you store Cluster information does not necessarily have to belong to any of the Clusters; it is just used to create and store the Cluster and Profile records.)


Be sure the network name for this server's C:\ root directory is mapped as a local drive on this computer.  For example, on FILESERVER, you would map N:\ as the C:\ (root) drive.


Go to Override Settings (under the Utilities menu); the option for Enable Multi-Directory Clustering must be on. (Exit out of ComQuest, then go back in after enabling this option, in order for the Station Clusters option under the Utilities menu to be displayed.)


Under the Utilities menu, select Station Clusters.  From here, add the name of each cluster, such as Rock Stations, Southwest Urban Stations, etc.



Station clusters browse


Next, to each cluster, add the various directories you want to be included in each specific cluster.



Adding a new database to a Cluster


There is no practical limit to how many directories can be linked to any one cluster, and a specific database can belong to more than one cluster, however it is not possible (nor should it be necessary) to include the same data directory (with the same tier) in a cluster more than once. You can, however, include different tiers from the same data directory.



Directories included in the Top 40 Cluster


Finally, add a new Report Profile.  This is how you will actually specify that a certain cluster is used when you run a report; you will choose the Report Profile that you have defined to represent a cluster of stations. Instead of designating specific stations, ethnicity or a Tier on the Report Profile, select the desired Station Cluster you want to be associated with this Profile.



Selecting Cluster in Report Profile form


For simplicity sake, you should name this new profile something similar to the name of the Cluster.  For example, if you have selected the Cluster Top 40 Stations, you might name the Profile Cluster: Top 40 Stations.


Now, when you run any individual Report, or when you run Batch Reports (from this database), simply select the Report Profile that matches the Station Cluster you wish to compile (such as the Cluster: Top 40 Stations Profile).  When the report is compiled (or for each report in your Batch), ComQuest will loop through all the databases that are associated with this Cluster, and compile one report, comprised of all available score data that meets your filter criteria.


Things To Consider


Keep in mind filter specs, such as date range, demo, gender, etc. will be observed uniformly across all databases.  Thus, if you run a Cluster report for "18-24 Females", but one of your databases has only a 17-25 cell, none of that directory's data will be included in the report (as it's not within the low/high age range you requested).


All song scores in each database that meet the date/age/gender criteria you designated when you ran the report will be included.  If the same songs are in all databases, this is no problem, but a song(s) that is in only one database will still be included in a Cluster report, although it obviously won't have as many scores as songs that are in all databases.  You might choose to include a Raw Counts report as one of the reports in a Cluster Batch, so that you can see the actual score count for each song.


Obviously, the more databases you have in a Cluster, the longer it will take each report you run using that Cluster to compile.  For example, a report using a Cluster consisting of 4 databases will take approximately 4 times as long to compile as a single-directory report.


Using a Station Cluster Report Profile prevents you from using another Profile for a report, but this should not be a problem as the other profiles are unique to a specific database (such as "WAAA P1's"), and this would not be relevant across multiple databases. (You can however select a specific Tier from a database, or leave at "0" to include all scores).


The Station Clustering feature is available for virtually all Song Score reports; although it is not available in any Perceptual or Cume/Pref reports, as these seem to be more market-specific, and would be less usable (or meaningful) in Clustering.


See also

Report Profiles