Interpreting Workflows
Learn how to interpret workflow results, understand key fields, and analyze outcomes in the FrankieOne KYC v2 API.
Overview of Results
After executing a workflow, the results provide comprehensive insights into the outcomes of the checks performed. Each workflow consists of multiple steps, such as Know Your Customer (KYC) checks, Anti-Money Laundering (AML) screenings, Identity Verification (IDV), and risk assessments. Each step autonomously documents its findings, contributing to the overall workflow result. To simplify reporting, step outcomes are grouped by category, ensuring that identical steps are aggregated and presented as a unified result. For example, KYC checks may be performed multiple times using various service providers and configurations, but the results are consolidated into a single step outcome that reflects the combined assessments. Below are the key fields and their significance:
Legend for Importance
The statusOverrideAt
field can be used to determine if the status was manually overridden.
Workflow Statuses & Results
Workflow execution statuses provide insights into the current state of the workflow. Workflow execution results provide insights into the final outcome of the workflow. Below are the possible statuses and results:
Workflow Step Results
Each workflow consists of multiple steps, each representing a specific check or process. Below are common steps and their possible outcomes:
Supplementary Steps
Start Step
The start step signifies the successful initiation of the workflow. To confirm that the workflow started without issues, check if the execution result includes a start step with a result
of COMPLETE
. Typically, no processResults
are expected for this step.
Decision Step
The decision step serves as a pivotal point in the workflow, determining its final status—PASS
, FAIL
, or REVIEW
. This decision is based on the outcomes of the preceding steps. For example, if all prior steps were executed successfully and all checks were cleared, the workflow is typically marked as PASS
. Conversely, if any step encountered an issue, such as a failed verification, the workflow is likely marked as FAIL
. It is important to note that these statuses can be customized within the workflow builder.
Risk Step
The risk step in the workflow is a mandatory checkpoint for risk assessment. It doesn’t result in a pass or fail status. Instead, it’s simply marked as “COMPLETED” once a risk evaluation has been conducted. This step ensures that risks are always considered in the process, without directly affecting the workflow’s status. While the risk assessment itself doesn’t change the workflow, the information gathered may be useful for later steps.
Finish Step
The finish step indicates whether the workflow has completed successfully. This means all results have been collected and stored, ready for further analysis. To verify that the workflow finished without issues, look for a result
of COMPLETE
. Note that this only confirms the workflow’s completion, not the success of individual checks or the absence of errors. Typically, no processResults
are generated for this step.
The table below represents the overall result of a workflow step. Each value provides a specific status or outcome of the step, depending on the context and its intended operation:
Service Provider Result
The KYC workflow step reflects the overall outcome from the service provider, represented by the serviceProviders.result
field.
Step Summaries
The workflow step summary provides a concise overview of the results of each step executed as part of the workflow. It can include one of the following summaries:
Each summary provides insights into the specific checks performed during the workflow execution, allowing organizations to assess the effectiveness of their KYC processes and make informed decisions based on the results. For more details on each step, refer to the public API documentation.
Process Results: Data Source Match Details
KYC match details are represented as processResults
existing under the workflow step results for steps that require verification or checks to be run. These details vary based on the context of what the check is trying to accomplish. For checks against external data sources, the processResults
aim to capture what exactly has matched against which source. Steps typically return an array of all the individual process results that were used in determining that step’s result
.
After a KYC step has been executed, various elements require examination to determine the final outcome.
Legend for Importance
Field Descriptions
When a process result object is first created, it is assigned an ID. When updating the process result object, make sure you set the processResultID to the ID of the process result object you are referring to
Example: KYC Data Source Match Details
Process Result Objects (PROs) encapsulate the outcomes of specific processes (check, comparison, verification etc.). They play a crucial role in evaluating the overall outcome of relevant workflow steps.
Let’s break down the following KYC process result:
- The
class
is “KYC” which indicates the process is in relation to a KYC database verification check. - The address of the entity has been successfully verified against the Australian Electoral Roll:
- Based on the
objectType
we can see that the specific element checked was an “ADDRESS” - The
providerResult
array contains the details of the service provider that the entity was checked against:- The service provider is the Document Verification Service as indicated by the
name
field (“dvs”) - The specific data source that was checked is the Australian Electoral Roll as indicated by the source (“au-elec-roll”) and
sourceNormalized
(“au-elec-roll”) fields
- The service provider is the Document Verification Service as indicated by the
- The result is “MATCH” indicating that the address was successfully matched.
- Based on the
- The check was completed successfully (
state
is “COMPLETED”) and hasn’t yet been invalidated in any way (systemStatus
is “VALID”), therefore it’s safe to determine that the process ran to completion and was indeed used in determining the successfulresult
. - There has been no manual status applied to the process (due to the absence of a
manualStatus
field)
Example: KYC Data Source Non-Match Details
Similar to Process Results (PROs) for KYC database verification matches, process results that result in no match will be generated during step execution and will provide insights into why the data source could NOT be matched.
Let’s break down the following KYC process result:
- The
class
is “KYC” which indicates the process is in relation to a KYC database verification check. - The address of the entity has been checked but wasn’t verified against the Australian Electoral Roll:
- Based on the
objectType
we can see that the specific element checked was an “NAME” - The
providerResult
array contains the details of the service provider that the entity was checked against:- The service provider is the Document Verification Service as indicated by the
name
field (“dvs”) - The specific data source that was checked is the Australian Electoral Roll as indicated by the source (“au-elec-roll”) and
sourceNormalized
(“au-elec-roll”) fields
- The service provider is the Document Verification Service as indicated by the
- The
result
is “MATCH” indicating that the address was successfully matched.
- Based on the
- The
matchStrengths
withinsupplementaryData
shows that no part of the name matched at all. - The fuzziness within
supplementaryData
shows that an exact match (normalized
as “0”) was required, potentially restricting any passable partial matching (“Sarah”, “Sara”). - The check was completed successfully (state is “COMPLETED”) and hasn’t yet been invalidated in any way (systemStatus is “VALID”), therefore it’s safe to determine that the process ran to completion and was indeed used in determining the successful
result
. - There has been no manual status applied to the process (due to the absence of a
manualStatus
field).
Example: AML Screening & Match Details
AML screening results are reflected in the result
field, which clearly indicates the steps outcome. This field summarizes the steps result, taking into account all the processed data. For instance, if an individual’s details such as name show up on any AML screening lists like PEP or Sanctions lists, the result will be HIT
. If not, the result will be CLEAR
.
processResults
exist under the workflow step results for steps that require verification or some sort of checks to be run and differ based on the context of what the check is trying to accomplish. For checks against external data sources, the processResults
will aim to capture what exactly has matched against what source. Steps typically return an array of all the individual process results that were used in determining that step’s result
.
Take for instance the following example:
KYC Matching Rule Results
The matchedRules
field offers a comprehensive breakdown of the rules applied during workflow execution. This field is instrumental in understanding how specific checks were conducted and which data sources were utilized to verify the individual’s information.
To delve deeper into the personal information matched against specific sources, refer to the matchedRules
. Typically, a single ruleMatches
entry outlines the outcome of a specific rule defined in the workflow.
For example, a rule might stipulate that both a name and either a date of birth or an address must be verified against two independent data sources. The matchTypes
field highlights the exact elements verified, while the matchSources
field identifies the data sources used for confirmation. Any data sources consulted but yielding no matches will be listed under nonMatchSources
.
Matched and Unmatched Rules
The matchedRules
and unmatchedRules
fields within the KYC workflow step summary share the same structure but differ in their outcomes. matchedRules
represents rules successfully matched, while unmatchedRules
represents rules that were not.
These rulesets are defined within the KYC workflow step configuration and determine whether the required number of data source matches have been met for each entity element (e.g., name, date of birth, address) for the verification to be considered successful.
By default, at least one matching rule record is provided at the conclusion of a KYC step, offering clarity on whether the verification criteria were satisfied.
Rule Details
Rule Matches
Summary of all KYC matches for a given set of criteria.
Match Types
The match types that this overall count and result refer to. Currently, one or more of the following:
- name
- address
- dob
- gender
These will be keys in a map whose values hold the values for the individual match types. For example:
KYC Example 1
Let’s break down the following match result:
name
,dob
andaddress
were checked as they’re listed undermatchTypes
- Based on the
matchCount
we can determine how many data sources we were able to match against for each element: name
matched against the Australian Electoral Roll (“au-elec-roll”) and Equifax Public Credit Data Header (“au-efax-cdh-pub”)address
matched against the Australian Electoral Roll (“au-elec-roll”) onlydateOfBirth
matched against the Equifax Public Credit Data Header (“au-efax-cdh-pub”)matchCount
equals thematchCountRequired
which means we were able to find enough matches for a positive resultrequiredSourcesMatched
indicates that the Australian Electoral Roll (“au-elec-roll”) was a source that was required to be matched against, which it was (name, address)isVerified
indicates that the match requirements were satisfied based on the rule criteria
KYC Example 2
Similarly to the matchedRules
, the unmatchedRules
contains a singular ruleMatches
entry detailing the outcome of a particular rule established in the workflow that hasn’t been met.
Let’s break down the following non matched result:
name
,dob
andaddress
were checked as they’re listed undermatchTypes
- Based on the
matchCount
we can see nomatchSources
present which tells us that there were no matches. - The
nonMatchSources
lists the sources that were checked against but without a successful match result name
didn’t match against the Australian Electoral Roll (“au-elec-roll”) and Equifax Public Credit Data Header (“au-efax-cdh-pub”) or Equifax Consumer Credit Data Header (“au-efax-cdh-consumer”)address
didn’t match against the Australian Electoral Roll (“au-elec-roll”) and Equifax Public Credit Data Header (“au-efax-cdh-pub”) or Equifax Consumer Credit Data Header (“au-efax-cdh-consumer”)dateOfBirth
didn’t match against the Australian Electoral Roll (“au-elec-roll”) and Equifax Public Credit Data Header (“au-efax-cdh-pub”) or Equifax Consumer Credit Data Header (“au-efax-cdh-consumer”)matchCount
doesn’t equal thematchCountRequired
which means we weren’t able to find enough matches for a positive resultrequiredSourcesMatched
indicates that the Australian Electoral Roll (“au-elec-roll”) was also a source that was required to be matched against, which wasn’tisVerified
indicates that the match requirements weren’t satisfied based on the rule criteria
Workflow Risk Assessment
The workflow risk assessment provides a comprehensive overview of the risk factors associated with the individual entity based on the checks performed during the workflow execution. It includes details such as:
Risk Factors
Each risk factor provides additional context about the risks identified during the workflow execution. Below are the details of each risk factor:
The status
field provides additional context about the risk factor:
VALID
: The risk factor is applicable for the entity.STALE
: The risk factor is no longer applicable for the entity.OVERRIDDEN
: The risk factor was manually overridden, and themanualOverrideScore
(if present) was used for the risk score calculation.DISCARDED
: The risk factor was added during the workflow execution but discarded later in the same workflow.
Best Practices
- Interpret Results Effectively: Thoroughly analyze workflow results to gain insights into the outcomes of the checks performed and identify any areas requiring attention.
- Proactively Handle Exceptions: Address issues or exceptions encountered during workflow execution promptly to ensure smooth operations.
- Ensure Workflow Finalization: Accurately record all results and take necessary actions based on the outcomes to complete the workflow process.
- Maintain Comprehensive Documentation: Keep detailed records of workflow execution, including manual overrides and exceptions, to support compliance and auditing requirements.
- Focus on Continuous Improvement: Regularly review workflows, gather feedback, and refine processes to enhance efficiency, accuracy, and effectiveness.
- Foster Collaboration: Partner closely with your implementation team to optimize workflows and ensure alignment with evolving compliance requirements and business goals.
- Adhere to Compliance Standards: Ensure all workflow executions and results management processes comply with applicable regulations and industry standards.
- Invest in Staff Training: Provide comprehensive training to staff involved in workflow execution and results management to equip them with the knowledge and skills to handle exceptions and overrides effectively.
- Implement Robust Monitoring: Establish monitoring mechanisms to track workflow performance, identify bottlenecks, and uncover opportunities for improvement.