Filtering Interactions by Metadata
Overview
The Interactions table allows you to efficiently locate and analyze specific interactions by filtering on contextual metadata. This capability helps you narrow down interactions based on information received from your integrations as well as insights automatically generated by Sedric AI.
Metadata-based filtering is especially useful for focusing reviews, applying the correct scorecard criteria, and identifying interactions that meet specific operational or compliance conditions.
Types of Metadata Available for Filtering
Sedric supports two types of metadata for each interaction. Both types can be used as filters in the Interactions table.
1. Metadata (Integration)
Integration metadata is received directly from your connected systems (such as dialers, CRMs, or other data sources) together with the interaction.
This metadata typically includes:
Phone number dialed
Client or Account ID
Call disposition or classification provided by the source system (for example: Do Not Call, Voicemail, etc.)
These values reflect the information as it was recorded at the time of the interaction by the external system.
2. Metadata (Sedric AI)
Sedric AI metadata is automatically generated by Sedric based on analysis of the interaction content.
This metadata:
Is derived from the conversation context
Can be customized to your organization’s specific needs
Provides higher-level insights and classifications
Common examples include:
Sedric’s classification of the call type
Detected hardship indicators
Detected disputes or regulatory signals
Note: Contextual metadata from Sedric AI is used to determine which scorecard criteria are applicable when evaluating each interaction.
Where to View Metadata in an Interaction
Metadata is displayed within each individual interaction page:
Purple chips at the top of the screen represent Sedric AI metadata
Clicking Show more reveals red/pink chips, which represent integration metadata
How to Filter Interactions by Metadata
To filter interactions using metadata:
Navigate to the Interactions table.
Click the + Filter button at the top of the page.
In the filter menu, select either:
Metadata (Integration), or
Metadata (Sedric AI)
Choose the relevant metadata key.
Select one or more values to apply the filter.
The selected metadata filter is applied immediately to the table.
Filter Logic and Behavior
Multiple Values for the Same Key (OR logic)
When you select multiple values for the same metadata key, the results will include any interaction that matches at least one of the selected values.
Example:
Call Type = Voicemail OR Do Not Call
The table will show all interactions that match either value.
Filtering on Keys with Many or Unique Values
For metadata keys that contain a large number of distinct values (such as Account ID):
The filter will display a free-text input instead of a long list
You can paste or type the exact value you are searching for
This allows precise filtering even for unique, interaction-level identifiers.
Multiple Keys (AND logic)
When you apply filters across multiple metadata keys, the results will include only interactions that match all selected keys, with at least one selected value per key.
Example:
Sedric AI metadata:
Disposition =RPCSedric AI metadata:
Dispute = Y
Only interactions where both conditions are met will be returned:
Metadata Filters as Table Columns
Each metadata filter you apply is also automatically added as a column in the Interactions table.
This makes it easier to:
Compare metadata values across interactions
Scan results visually without opening each interaction
You can:
Reorder metadata columns
Hide or show them using the Columns button
Summary
Filtering by metadata allows you to:
Quickly locate relevant interactions
Apply the correct scorecard criteria automatically
Focus reviews on interactions that meet specific operational or compliance conditions
By combining integration metadata and Sedric AI metadata, you gain flexible and powerful control over how interactions are filtered, reviewed, and analyzed.