Revisiting a SiteCatalyst Classic: Classification Fundamentals (SAINT)
by Brian Au
posted on 05-22-2013
Classifications (SAINT) are one of the enduring features of SiteCatalyst. SAINT (SiteCatalyst Attribute Import Naming Tool) allows for user-defined uploads of metadata within SiteCatalyst. The Adobe Marketing Cloud knowledge base outlines ‘A classification is a way of categorizing SiteCatalyst variable data, then displaying the data in different ways when you generate reports. When classifying, you are establishing a relationship between the variable and the metadata related to that variable.’ For this post, in advance of the upcoming Auto-Classifications Rule Builder feature release, I wanted to outline the overall value that Classifications can bring to an implementation and share key practical consideration areas. The main reference materials for Classifications on the Adobe Marketing Cloud Help Microsite can be found here: SiteCatalyst SAINT Classifications Home.
**Benefits of Classifications (SAINT) **
The benefits of Classifications are three-fold:
- Save & preserve key variables
- Simplify the implementation
- Apply metadata retroactively
Each of these areas of classifications can provide a lift in building a comprehensive, flexible, and scalable SiteCatalyst implementation. A key benefit to classifications is they do not require additional SiteCatalyst variables to be allocated to provide additional data within the reports. Classifications in other words can provide an organization with a multitude of additional views or slices of data; all derived from the same underlying captured source key value. A classic example of the value of classifications is the product hierarchy, whereby in SiteCatalyst only the product SKU is captured and all further related metadata is handled via classifications. Classifications therefore allow for a simplified approach to implementation data capture into variables, because not all data necessarily needs to be captured wholly within the SiteCatalyst image request. Another intrinsic aspect of classifications is the values are applied retroactively to a report which allows for later business decisions and revisions to be made after initial data capture of source key value.
**Practical Considerations **
Classifications carry several critical handling points:
- Importance of the Key Value – Users researching classifications will learn for many reporting issues it all comes down to the key value captured at the start. The source key value captured within the variable is the initial reference point for the entire classifications hierarchy and must meet the refinement and specificity requirements of an organization to functionally work. It is very important that duplicate assignment values are not issued for the same source key. The level at which the key can relate to end data also should be low enough in granularity to allow for further rollup via classifications.
- 1:1 Relationship with Classifications – All classifications always must exist on a 1:1 relationship with a source key value. It is not possible in SiteCatalyst classifications for a single source key value to contain multiple different classifications values for the same hierarchy level. If multiple values are uploaded within the same metadata upload file then the file will error out. Furthermore it is not possible for dynamic keys to exist as wildcards that relate to a single classification attribute level.
- User Upload Options – In terms of manual upload options, classifications data files can be uploaded by Admin users via the web browser or through FTP. The manual browser method steps flow through the Admin > SAINT Classifications menu in SiteCatalyst. The browser-based upload limits your SAINT data upload to a single report suite and there is a 1MB upload file size limit. Upload of classifications data files via FTP allows for multiple report suites at once and carries higher file size limits. The Adobe Marketing Cloud Help Microsite contains detailed instructions for both methods and scenario specific file size limit details Uploading SAINT Data Files, Upload A SAINT Data File via Web Browser, & Upload A SAINT Data File via FTP.
An important note is data dimensions that do not have an element value at the time the metric event happened are noted as an undefined ‘None’ value within the reports. A true ‘None’ in the pure sense is not applicable to the report on hand. However ‘None’ within a SAINT classification report hierarchy level could also mean keys have not been classified as well. Users need to be aware of the potential for the ‘None’ causes mix within their reports and act after thorough data checks on the variable.
**Other Classifications Considerations **
- Users cannot classify the number ‘0’ and must use a replacement such as the word ‘zero’
- SAINT classifications count against monthly unique limit 500k
- Characters length for all classifications is 255 bytes
- Total number of classifications should be limited to 30 classifications due to a hard limit on export of only the 30 classifications
- .TAB and .FIN extension upload files will not be processed and must be in lowercase
- Each variable needs to be classified separately within its own SAINT file
- The admin console is the only configuration location to create new classifications
**Classifications File Tricks **
- It is possible to classify the None bucket by entering classifications for the
nonekey. For example nonecan be set so that it covers Direct site traffic.
Classifications (SAINT) Review 1.1:
none Value File Action Example
- Users can “unclassify” prior uploaded metadata so that there is no value by entering the
emptyvalue in a given field.
Classifications (SAINT) Review 1.2:
empty Value File Action Example****
**General Classifications (SAINT) Guidelines **
- SAINT SiteCatalyst v15 uploads in 6 hour intervals, but it may take up to several days to manifest in the end reports depending on the scope of the upload. The time required to process a SAINT file varies on the size of the file and the current number of SAINT files already being processed by Adobe’s servers. Processing of SAINT files usually takes no longer than 72 hours.
- Upload of only the delta changes ensures most efficient metadata processing.
- SAINT Classifications can apply to all Custom Traffic s.props, s.pageName, s.channel, s.server, all Custom Conversion s.eVars, product name, campaign variable, video name.
- ****Logging out and logging back in may be required to show the proper update and refresh for any SAINT changes.
- The base source key value is always at the bottom of the SiteCatalyst menu flyout and the classifications above in the menu are the user specified metadata hierarchy levels.
Proper and savvy use of classifications is a critical component of a SiteCatalyst implementation. By leveraging what is already being captured, classifications enable a SiteCatalyst implementation to provide rich data views. Classifications provide organizational data flexibility within a user structured system. I hope the above post was a good refresher and provided additional clarity on all the valuable aspects of classifications.
Brian Au is a consultant in Adobe Consulting, focused on digital strategy, analytics & optimization for retail & travel clients. He tweets at _@_BrianAu.