Import Non-Speech Business Rules and Metadata
U-Analyze allows user to import call related attributes that are provided by the customer. User provides these attributes cross referenced with a Unique identifier like Call ID. These attributes are alpha numeric fields (with alphanumeric and spaces) and are very specific to business and expected to change based on project.
Customer can provide these attributes in any (input) format - XML, tab separated format, xls, CSV, etc. Implementation team transforms the input format into the import format (csv, comma separated) before importing into the system.
Call Attributes
Call Attributes can be of two types – Business Rules and Metadata
Business Rules (BR)
Business Rules are data which have low cardinality and are typically not unique for the calls. These can be treated as a dimension in the reports/filters and are associated with a business or product context.
Examples:
Zones - North, South, East, West
Card Type - Platinum, Gold, Silver
Like speech-related BRs are defined in U-Analyze, these imported BRs are also available in the Reports and Contact Analysis screens.
Metadata
These are data which have high cardinality and are mostly unique (not always) for the calls.
Examples:
Customer Phone Number, Customer Date of Birth
Imported Metadata is listed in the Call-wise Report as columns (after the BR values)
Template to Import Non-Speech Business Rules
Non-Speech BR template is in .csv format. Call Details of a call are listed in a single row.
Template for Non-Speech BR is saved as "<File Name>.nspbr"
Column Headers: Call ID, <BR1>, <BR2>, etc.
Header Example: Call ID, Zone, Card Type
Data Example1: C001, North, Platinum
Data Example2: C002, South, Platinum
For Multiple BR Values
Header Example: Call ID, Zone, Card Type, Call Type
Data Example3: C002, South, Platinum, Query|Complaint
![]() |
Template to Import Non-Speech Metadata
Non-Speech Metadata template is in .csv format.
Template for Non-Speech Metadata is saved as "<File Name>.nspmeta"
Column Headers: Call ID, <Metadata Type 1>, <Metadata Type 2>, etc.
Header Example: Call ID, Customer Phone Number, DOB
Data Example1: C001, 9873119110, 12/04/2000
Data Example2: C002, 8791199110, 13/05/1980
![]() |
Notes:
BRs and Metadata are saved as String type.
BRs and Metadata values can only be in English (no Unicode text)
In BRs Blanks are not allowed and they are considered as 0 in DB and whereas Metadata is considered as blank in DB
Pipe separators (for multiple values) are allowed for BRs
Values are case insensitive
Values are trimmed
BRs has the same validation as BR Hierarchy in GUI; Space, Letters, Numbers, Underscore are allowed
Metadata allows all characters.
System Validation
Warnings
Call ID should exist.
Call ID does now allow duplicates. (The first Call ID is imported)
For Standalone BRs, only 0 or 1 are allowed.
BR Header allow only its child BRs as values.
Errors
The BR names in the Header exist in the system as Non-Speech BR.
All Non-Speech Business Rules in U-Analyze must exist as headers in the Data sheet.
BR Names (part of BR hierarchy) and standalone BR are allowed as Headers in data sheet. Leaf BRs are not allowed as headers.
If warnings exist, System save the complete .csv file in "NonSpeech BR_Metadata Warnings". The records without warning are imported into the system. The warning message is displayed in the last column either "Call ID does not exist." or "Call ID already exists."
If errors exist, System save the complete .csv file in "NonSpeech BR_Metadata Errors". No records are imported.
Non-Speech BR and Metadata files is collected from the same Source folder as Audio Calls.
Templates are available as a part of Release package.
All data imports - meta data and BR of the calls from the current import overrides the previous data of those calls.
Notes
Additional Non-Speech BRs can be added for a new import.
If old calls are not reprocessed, then Non-Speech BRs are NULL.
CSV files with no Errors or Warnings are placed in "Non-Speech BR Metadata Success" folder.
Warning, Error, Success Folders are created at a Category level.
Calls which have only Non-Speech BRs cannot be processed. Score is calculated only through reprocessing when only Non-Speech BRs exist.
The data from a single or multiple CSVs is imported in a sequential manner.
System import the data only till before a blank row or blank column header. The data after a blank row or column header is not considered. Only the corresponding details of the imported data are referred in Success, Error or Warning folders.
If there is no value for a Business Rule, then the value of the BR is 0. If a Business Rule is evaluated to both 1 and 0, then the value of 1 is considered.
Example: A is a parent of B and C. B is a parent of D and E. C is a leaf node.
Data Sheet
||Call ID||A||B||
|C001|B|D|
|C002| |D|
|C003| | |
Values
||Call ID||A||B||C||D||E
|C001|1|1|0|1|0
|C002|0|1|0|1|0
|C003|0|0|0|0|0P
Additional Error Messages
Error Message in DB
Condition: When the uploaded CSV file is empty without any data (no header also)
Message: CSV File is empty.
Error Messages in CSV
Displayed on Second row in the last visible column
Call ID header is required in the first column.
Duplicate Column Names are not allowed. Call ID to be unique.
Few Business Rules configured in U-Analyze do not exist in the CSV File. The CSV File to contain all the configured Business Rules.
CSV File contains Business Rules that are not configured in U-Analyze.
At least one column name for Metadata is required.
Warning – On respective rows for which warning is applicable
Call ID does not exist.
Call ID already exists.
The values are invalid.