Tink.RiskCategorisation (Tink v0.1.1)

Risk Categorisation API for assessing financial risk profiles.

This module provides comprehensive risk assessment capabilities to categorize users based on their financial behavior and transaction patterns:

Spending behavior analysis
Financial risk scoring
Transaction pattern categorization
Risk indicators and flags
Behavioral risk assessment

Features

Risk Categories: Low, Medium, High, Very High risk classification
Behavioral Analysis: Spending patterns and financial habits
Risk Indicators: Specific risk flags (gambling, excessive spending, etc.)
Trend Analysis: Changes in financial behavior over time
Comprehensive Scoring: Multi-factor risk assessment

Flow

# Step 1: User completes Tink Link authentication
# (Build Tink Link URL in Console > Risk Categorisation > Tink Link)
# User authorizes and you receive risk_categorisation_id via redirect

# Step 2: Get access token
client = Tink.client(scope: "risk-categorisation:readonly")

# Step 3: Retrieve risk categorisation report
{:ok, report} = Tink.RiskCategorisation.get_report(client, risk_categorisation_id)

Use Cases

Loan Risk Assessment

def assess_loan_risk(client, risk_categorisation_id, loan_amount) do
  {:ok, report} = Tink.RiskCategorisation.get_report(client, risk_categorisation_id)

  risk_category = report["riskCategory"]
  risk_score = report["riskScore"]

  case risk_category do
    "LOW" ->
      {:approve, :low_risk}

    "MEDIUM" ->
      if loan_amount < 50000, do: {:approve, :medium_risk}, else: {:review, :medium_risk}

    "HIGH" ->
      {:review, :high_risk}

    "VERY_HIGH" ->
      {:reject, :very_high_risk}
  end
end

Credit Card Approval

def evaluate_credit_card_application(client, risk_categorisation_id) do
  {:ok, report} = Tink.RiskCategorisation.get_report(client, risk_categorisation_id)

  risk_flags = report["riskIndicators"] || []

  cond do
    "GAMBLING" in risk_flags ->
      {:reject, :gambling_detected}

    "EXCESSIVE_SPENDING" in risk_flags ->
      {:reject, :poor_financial_management}

    report["riskCategory"] in ["LOW", "MEDIUM"] ->
      {:approve, determine_credit_limit(report)}

    true ->
      {:reject, :high_risk}
  end
end

Account Monitoring

def monitor_account_risk(client, risk_categorisation_id) do
  {:ok, report} = Tink.RiskCategorisation.get_report(client, risk_categorisation_id)

  previous_category = get_previous_risk_category(risk_categorisation_id)

  if risk_increased?(previous_category, report["riskCategory"]) do
    notify_risk_team(%{
      user_id: report["userId"],
      old_risk: previous_category,
      new_risk: report["riskCategory"],
      indicators: report["riskIndicators"]
    })
  end

  :ok
end

Risk Categories

LOW: Minimal risk, stable financial behavior
MEDIUM: Moderate risk, some concerning patterns
HIGH: Significant risk, multiple warning signs
VERY_HIGH: Severe risk, immediate attention required

Risk Indicators

Common risk flags that may be present:

GAMBLING - Gambling-related transactions
EXCESSIVE_SPENDING - Spending beyond means
OVERDRAFTS - Frequent overdraft usage
PAYDAY_LOANS - Use of payday lending services
DEBT_COLLECTION - Debt collection activity
RETURNED_PAYMENTS - Failed payment attempts
IRREGULAR_INCOME - Inconsistent income patterns
HIGH_CREDIT_UTILIZATION - Maxing out credit limits

Required Scope

risk-categorisation:readonly

Summary

Functions

get_report(client, report_id)

Retrieves a Risk Categorisation report.

Functions

get_report(client, report_id)

@spec get_report(Tink.Client.t(), String.t()) ::
  {:ok, map()} | {:error, Tink.Error.t()}

Retrieves a Risk Categorisation report.

After the user completes authentication through Tink Link, you receive a risk_categorisation_id. Use this ID to retrieve a comprehensive risk assessment report.

Parameters

client - Tink client with risk-categorisation:readonly scope
report_id - Risk categorisation ID (received via redirect after user auth)

Returns

{:ok, report} - Complete risk categorisation report
{:error, error} - If the request fails

Examples

# After user completes Tink Link flow:
# https://yourapp.com/callback?risk_categorisation_id=risk_cat_abc123

client = Tink.client(scope: "risk-categorisation:readonly")

{:ok, report} = Tink.RiskCategorisation.get_report(client, "risk_cat_abc123")
#=> {:ok, %{
#     "id" => "risk_cat_abc123",
#     "userId" => "user_123",
#     "status" => "COMPLETED",
#     "createdAt" => "2024-01-15T10:00:00Z",
#     "riskCategory" => "MEDIUM",
#     "riskScore" => 45.5,
#     "riskIndicators" => [
#       "OVERDRAFTS",
#       "HIGH_CREDIT_UTILIZATION"
#     ],
#     "analysisPeriod" => %{
#       "start" => "2023-01-01",
#       "end" => "2024-01-15",
#       "months" => 12
#     },
#     "spendingBehavior" => %{
#       "averageMonthlySpending" => %{
#         "amount" => %{"value" => 28000.0, "currencyCode" => "SEK"}
#       },
#       "spendingVariability" => "MODERATE",
#       "essentialSpendingRatio" => 0.65,
#       "discretionarySpendingRatio" => 0.35
#     },
#     "incomeStability" => %{
#       "stabilityScore" => 0.82,
#       "regularity" => "REGULAR",
#       "averageMonthlyIncome" => %{
#         "amount" => %{"value" => 35000.0, "currencyCode" => "SEK"}
#       }
#     },
#     "creditBehavior" => %{
#       "overdraftFrequency" => 3,
#       "averageOverdraftAmount" => %{
#         "amount" => %{"value" => 2500.0, "currencyCode" => "SEK"}
#       },
#       "creditUtilizationRatio" => 0.72,
#       "paymentHistory" => "MOSTLY_ON_TIME"
#     },
#     "riskFactors" => [
#       %{
#         "type" => "OVERDRAFTS",
#         "severity" => "MEDIUM",
#         "frequency" => 3,
#         "impact" => "Frequent overdraft usage indicates cash flow issues"
#       },
#       %{
#         "type" => "HIGH_CREDIT_UTILIZATION",
#         "severity" => "MEDIUM",
#         "value" => 0.72,
#         "impact" => "High credit utilization may indicate financial stress"
#       }
#     ],
#     "categoryBreakdown" => %{
#       "gambling" => %{
#         "monthlyAverage" => %{"value" => 0.0, "currencyCode" => "SEK"},
#         "percentageOfSpending" => 0.0
#       },
#       "paydayLoans" => %{
#         "detected" => false,
#         "count" => 0
#       },
#       "debtCollection" => %{
#         "detected" => false,
#         "count" => 0
#       }
#     },
#     "recommendation" => "MODERATE_RISK",
#     "confidence" => "HIGH"
#   }}

Report Structure

Risk Assessment

riskCategory: Overall risk classification (LOW/MEDIUM/HIGH/VERY_HIGH)
riskScore: Numerical score (0-100, higher = more risk)
riskIndicators: List of detected risk flags
confidence: Assessment confidence level

Spending Behavior

averageMonthlySpending: Average spending amount
spendingVariability: LOW, MODERATE, HIGH
essentialSpendingRatio: Percentage on necessities
discretionarySpendingRatio: Percentage on non-essentials

Income Stability

stabilityScore: Income consistency (0-1)
regularity: REGULAR, IRREGULAR, VARIABLE
averageMonthlyIncome: Average income amount

Credit Behavior

overdraftFrequency: Number of overdrafts in period
creditUtilizationRatio: Percentage of credit limit used
paymentHistory: ON_TIME, MOSTLY_ON_TIME, LATE, VERY_LATE

Risk Factors

Detailed breakdown of each risk factor:

Type and severity
Frequency or value
Impact description

Category Breakdown

Gambling activity
Payday loan usage
Debt collection presence
Other high-risk categories

Use Cases

# Check for specific risk indicators
{:ok, report} = Tink.RiskCategorisation.get_report(client, report_id)

if "GAMBLING" in (report["riskIndicators"] || []) do
  flag_for_manual_review()
end

# Calculate risk-based pricing
risk_score = report["riskScore"]

interest_rate = case risk_score do
  score when score < 25 -> 3.5  # Low risk
  score when score < 50 -> 5.5  # Medium risk
  score when score < 75 -> 8.5  # High risk
  _ -> nil  # Reject
end

# Monitor credit utilization
utilization = get_in(report, ["creditBehavior", "creditUtilizationRatio"])

if utilization > 0.8 do
  send_credit_limit_warning()
end

# Check income stability
stability = get_in(report, ["incomeStability", "stabilityScore"])

if stability < 0.6 do
  require_additional_verification()
end

Required Scope

risk-categorisation:readonly

Tink.RiskCategorisation (Tink v0.1.1)

Features

Flow

Use Cases

Loan Risk Assessment

Credit Card Approval

Account Monitoring

Risk Categories

Risk Indicators

Required Scope

Links

Summary

Functions

Functions

get_report(client, report_id)

Parameters

Returns

Examples

Report Structure

Risk Assessment

Spending Behavior

Income Stability

Credit Behavior

Risk Factors

Category Breakdown

Use Cases

Required Scope