An overview of risk insights
Wondering what you can get from our Risk Insights? Well, wonder no more!
Building a credit model demands a meticulous analysis of data, variable calculation, and data cleaning. To help data science teams get straight to building their models, our Risk Insights resource offers a comprehensive list of cleaned and pre-calculated features. With these features, data scientists have the building blocks to quickly create, iterate, and validate their models.
To calculate the risk insights, we take up to 365 days of transactional data from the user's checking, savings, investment, loan, and credit card accounts.
In this guide, we'll give you a quick overview of the main metrics that your teams can take advantage of to boost their models.
Asset Metrics
Asset metrics can give a snapshot of your user's wealth and liquid assets, which can indicate how they manage their money as well as their current financial status.
These metrics can be useful for:
- Developing customer profiles or wealth segments.
- Predicting future financial behavior and investment potential.
- Personalizing financial advice or product recommendations.
Credit Card Metrics
Credit card metrics are currently not available for EYOD Risk Insights.
Credit card metrics can give you a great indication of your customer's credit card habits. This includes how many credit card accounts a customer has, their total credit limit, how much of that limit they've already used, and the rate of their credit card limit utilization.
These metrics can be useful for:
- Checking if your customer is a good candidate for credit (creditworthiness and risk).
- Understanding how your customer uses their credit card (credit card spending habits)
- Evaluating whether to offer additional credit products or increase credit limits.
- Identifying early signs of financial trouble, especially if there is high credit card limit utilization.
Loan metrics
Loan metrics are currently not available for EYOD Risk Insights.
Loan metrics can help you understand your customer's borrowing and repayment behavior, which you can take into consideration when assessing their ability to take on additional credit and identifying potential default risks. It includes information on the number of loan accounts, the total and outstanding principal of these loans, and monthly payments. In addition, we also include the checking account overdraft limit and utilization (if available for the institution).
These metrics can be useful for:
- Assessing the customer's debt burden and their ability to take on more credit.
- Evaluating the reliability of the customer in terms of loan repayments.
- Forecasting potential default risks.
- Identifying opportunities for refinancing or offering additional loan products.
Using the asset, credit, and loan metrics are a great basis to create a customer profile.
Balance Metrics
Balance metrics are a great way to get a picture of your customer's overall financial health by looking at their aggregate account balance over time.
This can be a useful indicator as to whether your customer typically maintains a high balance (potentially an opportunity for upselling or other product recommendations) or if they often deplete their funds (which can be an indicator of financial trouble).
These metrics can be useful for:
- Checking your customer's overall financial health and stability.
- Identifying customers that may be at risk of overdrafts or financial troubles.
- Creating alerts or interventions for customers with consistently low balances.
Over time, changes in these metrics can highlight potential financial difficulties, opportunities for personalized product recommendations, and other insights to guide user engagement and support strategies.
| Metric group | Description |
|---|---|
| Minimum | The lowest recorded aggregate account balance over different time periods. |
| Mean | The mean (or average) aggregate account balance provides insight into the typical balance a user maintains over various periods. |
| Maximum | The highest recorded aggregate account balance over different time periods. |
| Standard Deviation | Provides insight into the variability or dispersion of a user's balance over time. A high standard deviation indicates significant fluctuation in the user's account balance, while a low value suggests the user's balance is relatively stable. |
| Trend | The ‘directional’ movement of the user's aggregate account balance over different periods. These metrics can indicate whether the user's balance is generally increasing, decreasing, or remaining constant. |
| Days Balance Below 0 | Counts the number of days when the user's balance drops below zero. Frequent occurrences might indicate a user struggling financially or failing to manage their account effectively. |
| Days Balance Below Mean | Counts the number of days when the user's balance falls below their average balance. This can provide an additional dimension to understand a user's balance management. |
| Days Balance Below threshold | The number of days the user's balance falls below a predefined threshold ("balance_threshold_x"). Consistently falling below this threshold can be an early warning sign of financial distress. |
Calculated periods
For all balance metrics, we calculate data for the following intervals:
- three days
- one week
- one month
- three months
- six months
- twelve months
Transaction Metrics
Transaction metrics give you a granular view of your customer's spending patterns (transactionality). They provide you with the frequency, volume, and type (incoming or outgoing) of transactions.
These metrics can be useful for:
- Understanding a customer's financial habits.
- Identifying changes in income or spending that could indicate life events or financial trouble.
- Grouping customers with similar spending habits for targeted marketing or product recommendations.
- Building models to predict future transactions and understand the customer's short-term financial needs.
| Metric group | Description |
|---|---|
| Overall Transactions Metrics | The total, mean (average), and maximum number of transactions over different time periods. Can be an insightful overview of user engagement by detailing how frequently users transact, their average transaction rate, and the maximum number of transactions. |
| Incoming Transactions Metrics | - The total, mean, and maximum number of incoming transactions (transactions where your user is the recipient). - The total, mean, and maximum sums of incoming transactions. |
| Outgoing Transactions Metrics | - The total, mean, and maximum number of outgoing transactions (transactions where your user is the payer). - The total, mean, and maximum sums of outgoing transactions. |
| Days without transactions | The number of days your user did not engage in any transactions, which can be useful in determining their activity. |
| Days since the last transaction | The number of days since the user's last transaction, which is useful for determining their activity or if this account is their current account. We also separately provide you with the days since the last incoming or outgoing transaction. |
| Days history | Total number of days for which transactional data is available for the user. |
Calculated periods
For all transaction metrics, we calculate data for the following intervals:
- three days
- one week
- one month
- three months
- six months
- twelve months
Cash flow metrics
Cash flow metrics can help in understanding the actual movement of money in and out of a customer's institution. They can provide insights into your customer's liquidity and their ability to manage their financial obligations.
These metrics can be useful for:
- Understanding the customer's financial health and liquidity.
- Assessing the stability of income and expenses.
- Identifying trends or patterns in the customer's cash flow.
- Predicting future cash flow situations, which can be useful for offering timely financial products or advice.
| Metric group | Description |
|---|---|
| Maximum positive and negative cash flows | The maximum amount of money coming into (positive) and leaving (negative) the user’s accounts in the institution. |
| Mean positive and negative cash flows | The mean amount of money coming into and leaving the user’s accounts in the institution. |
| Sum of positive and negative cash flows | The total sum of money coming into and leaving the user’s accounts in the institution. |
| Trends of positive and negative cash flows | The trend of money coming into or leaving the user’s accounts in the institution. |
| Positive-to-negative cash flows | A comparison of the user's positive and negative cash flows. A ratio greater than 1 suggests the user's income exceeds their spending, while a ratio less than 1 means the user spends more than they receive. |
Calculated periods
For all cashflow metrics, we calculate data for the following intervals:
- three days
- one week
- one month
- three months
- six months
- twelve months
Category metrics
Category metrics offer a granular view into your customer’s main spending patterns (the composition and nature of their day-to-day transactions). We currently return 15 primary categories and 94 detailed sub-categories. With these metrics, you can quickly view, per category, how much your customer spends in daily life, how much of this constitutes their total spending, and whether the pattern is trending upwards and downwards.
These metrics can be useful for:
- Understanding your customer’s financial habits.
- Offering personalized credit products and solutions.
- Building models to predict future transactions and understand your customer’s short-term financial needs.
For details regarding the categories (and sub categories), see our Categorizing transactions section.
| Metric group | Description |
|---|---|
| Net amount | The net amount of the transactions for this category in the last three months (calculated as the total incoming - total outgoing transactions for this category). |
| Inflow ratio | How much does this category (and subcategory combination) contribute to all the inflow transactions the user has made. |
| Trend | The trend of the net amount for the category (and subcategory) combination. |
Calculated periods
For all category metrics, we currently only calculate data for an interval of three months.
Updated 9 months ago
Want to see the full list of Risk Insight metrics? Check out our API reference: