Defining primary metrics and KPIs is essential for any data-driven organization. These metrics help us track our progress toward our goals and make better decisions. Each metric has its own formula for calculation, and each formula has a data source with the relevant data.
For example, if we aim to increase sales, we might define a primary metric as “Total Sales.” The formula for total sales is “number of units sold * price per unit.” The data source for total sales is our sales database or CRM system.
With a single source of truth for metric formulas, business definitions, query logic, and data origins, you can now implement governance to prevent cases where two departments in your company have two different answers to the same question. This type of augmented governance makes it easier to implement a data mesh architecture.
There are two ways to create your metrics in illumex:
- Define a formula over the auto-generated business terms (the same way you would do it in Excel or Google Sheets), and add a free text name and description to it. That’s all! The metric gets an auto-generated SQL query with a single definition of its SELECT statement with the formula calculation, FROM statement with a certain table (or a result of joined tables), and WHERE clause with the applied logic and filters.
- Choose your metric from the list of auto-suggested metrics constructed by illumex based on your usage patterns and industry standards, with SQL query and formula over business terms included.
Regardless of how you maintain your metric store, illumex’s AI Assistant suggests the name and description to streamline the process further.
For either of the mentioned approaches, illumex can also identify common partitions and filters frequently used with the metric. Scanning your usage patterns can also provide insights into the misuse and conflicts of certified metrics.
With a single source of truth for metric formulas, business definitions, query logic, and data origins, you can now implement governance to prevent cases where two departments in your company have two different answers to the same question. This type of augmented governance makes it easier to implement a data fabric or data mesh architecture.