When a caller reaches your contact center and the call is assigned to an agent, your agent can see, in the “Identity” tab in Conversations, the caller’s available data in real time.
The presented information to the agents is immediately available within the Identity app itself, where admins will be able to compare risk scores and decide which actions they, or their agents, need to take.
To access Identity, please select the Identity icon on the leftmost side panel, and you see the list of contacts that have used Voice authentication or had a background check performed on their phone number.
- “Name” [1]: Contact’s name and/or phone number, and ID, if another identifier besides the caller’s phone number was requested from the caller.
-
“Call Risk” [2]: The risk is based on different risks and fraud scores associated with the call, which are analyzed to reach a ranked risk score (from 0% to 100%) from which a chart is derived to represent criticality:
- Green: from 0% to 49%.
- Yellow: up to 70%.
- Red: above 70%
-
“Voice Authentication” [3]: Shows the result of the caller’s authentication:
- “Success” - If the contact’s voice biometrics matches the contact’s voiceprint.
- “Failure” - In case there’s no match of the voiceprint.
-
“Phone Validation” [4]: The phone validation fraud score:
- “High-risk”: Fraud scores >=85% are high-risk users that are likely to engage in malicious behavior. Scores in this threshold indicate recent or excessive abuse and fit the profile of a typical risky user.
- “Suspicious”: Fraud scores >= 75% are suspicious and are likely to be involved in suspicious or dangerous activities, but not necessarily a fraudulent user.
- “Safe”: Fraud scores < 75% are considered safe.
-
“Spoofing detection” [5]: The spoofing detection status:
- “High-risk”: The initiating service provider has authenticated that it placed the call on its network, but has no relationship with the caller.
- “Suspicious”: The carrier can authenticate its relationship with the calling customers, but cannot verify if the consumer is authorized to use the number.
- “Safe”: The service provider has fully authenticated that the customer initiating the call is authorized to use the calling number.
- “When” [6]: The date and time when the call took place.
Viewing a contact’s details
By clicking on any contact’s name, the corresponding details will open. Here, you see the same information you can find on the Identity list for the contact’s last call.
You can also browse the history and corresponding detailed information of the risk score of the previous calls, as well as the Voice authentication, Phone validation, and Spoofing detection results.
Managing Reporting
The Export button enables the download of identity data from interactions into a CSV file.
After clicking the Export button and before the download begins, filters can be applied to extract precisely the information needed.
Filters include options to specify a period of time, either by choosing predefined ranges* or selecting custom date ranges. Additionally, more granular filters are available, such as Interaction ID (allowing multiple IDs separated by commas), Voice Operation, Voice Result, Phone Status, Phone Valid, and Spoofing Status.
*Note: When using the Period of Time filter with custom date ranges, data can only be exported in segments of up to 180 days at a time.
Filtering is especially useful when dealing with large volumes of data. It reduces the risk of long download times or potential timeouts by enabling the extraction of smaller, more focused data samples that exclude unnecessary information.
After downloading, the CSV file will contain rows representing actions performed during specific calls. Each row represents an action, such as enrollment in active or passive voice authentication or a phone validation.
The CSV file includes several columns, organized as follows:
- Caller Information: Columns like contact_id (explained in the Contact ID FAQ) and phone_number provide details about the caller.
- Phone Validation Results: Columns beginning with phone_ indicate the outcome of the phone validation process, such as phone_status, phone_valid, phone_country, and more.
- Spoofing Detection: The phone_call_attestation columns provide results related to spoofing detection.
- Voice Authentication and Enrollment: Columns starting with “voice_” display results of enrollment and authentication, such as voice_operation, voice_risk_score, voice_result, and others.
- Call Summary: Columns such as interaction_id, timestamp, and overall_riskscore provide overall information about the call.
To effectively interpret the data in the CSV file, it is possible to filter by contact_id and then sort by timestamp. This approach organizes the voice authentication and phone validation details for a specific caller in chronological order.
It is possible to then identify rows containing enrollments and authentications associated with the same call by grouping them under the same interaction_id.
To help illustrate this process, two sample extracts from a testing environment are provided:
- For the contact ID 8965
- For the contact ID +16413292642
As detailed in the Contact ID FAQ, the contact_id can either be the contact’s phone number or a unique identifier.
Let’s start by analyzing the example for the contact with ID 8965:
- All columns starting with phone_ have the value NA (Not Applicable), indicating that no phone validation was performed for this caller. This occurs because the number dialed by the contact is linked to a Studio flow without the Validate Phone component configured.
- The first row indicates that this contact has consented to voice authentication. This is evident from the interaction_id column, which starts with the string "CONSENT," followed by an encrypted identifier that does not represent a call interaction ID.
- Additionally, the first row shows the date and time when consent was provided in the timestamp column. The voice_operation displays "CONSENT," and the voice_result shows "ALLOWED," confirming that the caller agreed to enroll in voice authentication. Following this, the agent proceeded by clicking the Enroll button.
- In the second row, the voice_operation is "ENROLL," and the voice_result is "SUCCESS," indicating that the caller successfully enrolled in voice authentication at the date and time specified in the timestamp column of this row. Furthermore, the voice_type column reveals that the caller enrolled in Active voice authentication.
- The third row records a successful authentication. The voice_operation is "VERIFY," and the voice_result is "SUCCESS," showing that the caller successfully authenticated. The voice_risk_score is 0%, indicating that the caller’s voice was an exact match with the voiceprint collected during enrollment. The overall_riskscore column in the same row also displays a value of 0%.
The second example focuses on the contact with ID +16413292642.
- Although all rows share the same date, "2022-11-23," they correspond to three separate calls, as indicated by three distinct interaction_id values.
- In the first row, only phone validation was performed, evident from the populated values in the phone_ columns. Meanwhile, the voice_ columns, related to voice authentication, have the value NA. The call risk for this row is 79%, as shown in the overall_riskscore column.
- The second row indicates that the caller gave consent to enroll in voice authentication. This is reflected in the interaction_id column, which begins with "CONS
- Rows 3, 4, and 5 belong to the same call, as they share the same interaction_id.
- Rows 3 and 4 show that the caller successfully enrolled in and authenticated using passive voice authentication.
- In row 4, both voice authentication results and phone validation outcomes are provided, together with the call risk score, shown in the overall_riskscore.
- Row 5 shows that the caller successfully enrolled in active voice authentication. Active voice authentication enrollment typically occurs when the agent ends the call, and the caller completes the process via the IVR system. This is why the interaction_id remains the same as in rows 3 and 4.
The 6th row corresponds to a different call, as indicated by a new interaction_id. In this call, the caller was successfully authenticated using active voice authentication. The row also includes phone validation results and the call risk, which is 37.5% in this instance.
The following table describes the fields exported in the CSV file:
|
Column name |
Description |
Example |
|
contact_id |
The contact identifier.
To know more check Contact ID FAQ. |
142123 |
|
interaction_id |
The ID of a call.
There will be several rows with the same interaction_id that represent what happened during a single call.
Note.
When the customer agrees to using voice authentication, this is called “Consent”.
When a customer opts out of voice authentication, their voiceprint is deleted.
To know more, read our FAQ. |
Examples of possible values: • c5572384a4c340c2b326965fdb13ca0e • CONSENT_388d0f2d-5112-4654-99ac-ce2b5b5af768 • DELETE_0111e6e3-7856-49b4-a501-9bfd3f307885 |
|
timestamp |
Date and time the action occurred. |
2020-02-12T10:00:00.000Z |
|
phone_status |
The phone validation fraud score
To know more, check Phone Validation Overview |
One of the following: • "HIGH_RISK" • "SAFE" • "SUSPICIOUS" • "ERROR" |
|
phone_formatted |
The phone number is formatted in the international dialing code. “NA” if not formattable. |
+13345555555 |
|
phone_local_format |
The formatted phone number in the country's local routing rules with area code. “NA” if not formattable. |
(334) 555-5555 |
|
phone_valid |
Indicates if the phone number is considered valid based on assigned phone numbers available to carriers in the caller’s country. |
One of the following: • TRUE • FALSE • NA |
|
phone_risk_score |
The risk score, measures how likely a phone number is to be fraudulent. |
A percentage from 0 to 100 with no decimal numbers • Fraud scores < 75% are considered Safe • Fraud scores >= 75% and < 85% are Suspicious • Fraud scores >=85% are High-risk |
|
phone_recent_abuse |
Details whether this phone number has been associated with recent or ongoing fraud. |
One of the following: • TRUE • FALSE • NA |
|
phone_voip |
Illustrates if this phone number is a Voice Over Internet Protocol (VOIP) or a digital phone number. |
One of the following: • TRUE • FALSE • NA |
|
phone_prepaid |
Indicates if this phone number is associated with a prepaid service plan. |
One of the following: • TRUE • FALSE • NA |
|
phone_risky |
Points out if this phone number is associated with fraudulent activity, scams, fake accounts, or other suspicious behavior. |
One of the following: • TRUE • FALSE • NA |
|
phone_active |
Reveals if this phone number is a live phone number that is currently active. |
One of the following: • TRUE • FALSE • NA |
|
phone_carrier |
The carrier this phone number has been assigned to, or "NA" if unknown. |
IRISTEL INC. |
|
phone_line_type |
The type of line this phone number is associated with or "NA" if unknown. |
One of the following: • VOIP” • “Toll-Free” • “Wireless” • “Landline” • “Satellite” • “Premium Rate” • “Pager” • “NA” |
|
phone_country |
The two-character country code for the phone number. |
Refer to “What is the country code mapping?” for a complete list of codes. |
|
phone_region |
The region (state) of the phone number if available, or “" if unknown. |
AL |
|
phone_dialing_code |
The 1 to 4-digit dialing code for this phone number or null if unknown. |
1 |
|
phone_call_attestation_type |
The call attestation level: • Full attestation - A: The highest level, meaning that the caller is known and has the right to use the phone. number that is making that call. • Partial attestation - B: The second level, where the call is from a customer known to this provider but the right to use the phone number cannot be verified. • Gateway attestation - C: This is the lowest level, where neither the caller nor the calling number can be authenticated, but the call is going through a known gateway.
For more information, read this article. |
On of the following: • A • B • C • NA |
|
phone_call_attestation_valid |
Besides the call attestation, the final carrier needs to confirm that the attestation was the one presented to them.
This field presents this confirmation.
To know more, read our FAQ. |
One of the following: • TRUE • FALSE • NA (if the phone_callattestation_type is NA) |
|
voice_operation |
Indicates the type of voice authentication operation that was performed: • "VERIFY": An active or passive authentication was performed • "DELETE": The customer has opted out of voice authentication • "ENROLL": An active or passive enrollment was done • "CONSENT”: The customer has consented to voice authentication |
One of the following: • "VERIFY" • "DELETE" • "ENROLL” • "CONSENT" |
|
voice_risk_score |
The caller's voice is analyzed to reach a ranked risk score (from 0% to 100%).
Note: Sometimes when an error occurs during authentication, that is, when “voice_operation” is VERIFY and “voice_result” is ERROR, the voice_risk_score is -1. This means that the system was not able to evaluate the authentication and didn’t receive a score. It may happen in cases when the caller hangs up before completing the authentication or a problem occurred during the evaluation that prevented the system to return the final result. |
A percentage from 0 to 100 with decimal numbers
Example: 33.99999738 |
|
voice_result |
This column is closely tied to the voice_operation column.
It represents the result of the operation performed: • SUCCESS ◦ If the voice_operation is VERIFY, the contact’s voice biometrics matches the contact’s voiceprint. ◦ If the voice_operation is DELETE, the contact voiceprint was deleted ◦ If the voice_operation is ENROLL, the customer voiceprint was created • FAILURE ◦ If the voice_operation is VERIFY, the contact’s voice biometrics doesn’t match the contact’s voiceprint. ◦ If the voice_operation is DELETE, the contact voiceprint could not be deleted ◦ If the voice_operation is ENROLL, the customer failed to enroll and their voiceprint was not created • ERROR - If an error occurred. • TIMEOUT - If the maximum waiting time is reached before the operation is completed • NO_SPEECH - (only if voice_operation is VERIFY) If the caller didn't speak. • UTTERANCE_FAILURE - (only if voice_operation is VERIFY) If the caller tried to authenticate with a different passphrase from the one provided. • DEEP_FAKE - (only if voice_operation is VERIFY) If the caller used a deep fake playback of a person's voice. • REPLAY_ATTACK - (only if voice_operation is VERIFY) If the caller used a recording device to play back another person's voice. • ALLOWED - (only if voice_operation is CONSENT) if the customer consented to enroll in voice authentication • REFUSED - (only if voice_operation is CONSENT) if the customer refused to enroll in voice authentication. |
|
|
voice_failure_code |
When a voice enrollment or authentication fails, a failure code and failure reason are generated, which can be valuable for troubleshooting. This code is particularly useful in expediting investigations conducted by the Talkdesk team, if needed. |
2003 |
| voice_failure_reason |
In the event of a voice enrollment or authentication failure, a failure code and failure reason are generated, providing valuable insights for troubleshooting. The failure reason explains the cause of the failure, enabling users to take appropriate action based on the specific issue. For more information about Voice Biometric Failures, please consult the Troubleshooting Voice Biometric Failures. |
UTTERANCE_FAILURE |
|
overall_riskscore |
The overall risk score, aka Call Risk, is an average that is calculated based on the several scores associated with the call.
Note: If the only score evaluated in the call was voice authentication that errored, then the overall risk score will have the same -1 value as the voice risk score, as explained in the “voice_risk_score”. |
A percentage from 0 to 100 with decimal numbers
Example: 37.5 |
|
phone_number |
The phone number without any formatting. |
+13345555555 |
|
voice_type |
The voice authentication type: • Active. • Passive.
When the customer consents to enroll in voice authentication or opts out from it, this operation is independent of the voice authentication type, so the column will show NA.
To know more about the difference between voice authentication types, read here. |
One of the following: • ACTIVE. • PASSIVE. |
|
name |
The value of this column is always “NA”. Due to a threshold imposed on the CSV export in the Identity app, the contact name cannot be retrieved. You can either:
|
NA |
To know more about Identity, please consult our documentation.