> For the complete documentation index, see [llms.txt](https://docs.gaiodataos.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.gaiodataos.com/tools/discovery/insights.md).

# Insights

<figure><img src="/files/mPNpjWHqGt7Iy5rPDSow" alt=""><figcaption></figcaption></figure>

The **Insights** task in the Discovery module is designed to help users find trends of growth, decline and abrupt changes in your indicators . It is organized into three subtabs: **General**, **Advanced**, and **Schedule**.

***

## **How to Use**

***

### **1. Access the Insights task**

* In the **Studio**, go to the Delivery panel.
* Click on plus button.
* Click on insights botton.
* Go to insights tab.

***

### **2. Configure the** General Tab

Use this tab to define the analytical structure that will guide the automatic insights engine:

* **Dimensions**: Select the categorical variables (e.g., Region, City) to break down the analysis.
* **Measure Type**: Choose how the metrics will be evaluated. Options include:
  * *Field*: direct field analysis
  * *Cumputed*: custom metric logic
* **Measures**: Choose the numeric fields to analyze (e.g., sales, profit).
* **Dates**: Select the date field to enable time-based analysis.

{% hint style="warning" %}
**Tip**: Always prioritize dimension hierarchy (e.g., Country → State → City) to improve the granularity of the insights.
{% endhint %}

***

### **3.Configure the Advanced Tab**

Customize the insight generation formatting:

* **Default Period**: Choose the default time granularity for trends (e.g., Dayly, Weekly, Monthly).
* **Growth Percentage (%)**: Set a growth threshold to highlight significant changes.
* **Percent/Decimal Size**: Define how many decimal places to display for percentage and numeric metrics.
* **Separator**: Choose between `dotComma` or `commaDot` for number formatting (e.g., 1,000.00 vs 1.000,00).
* **Invert Colors**: Enable this to flip the color scheme, useful for emphasizing negative trends.

***

### **4. Configure the** Schedule Tab

Use this section to automate the refresh of your insights periodically:

* **Every**: Define the frequency (e.g., Minute, Hour, Day, week, month, Year).
* **Time Options**:
  * *Of the month*: Choose specific days of the month.
  * *On the day / weekday*: Define specific calendar days or weekdays.
  * *At hour / minute*: Set precise execution times.
* **Status**: Toggle the task as *Active* to enable scheduling.

{% hint style="warning" %}
**Note**: Changes will only apply to future runs. Ongoing jobs will not be interrupted.
{% endhint %}

***

### **5. Save the Configuration**

* Click **Save** in the top-right corner to apply your settings.

***

### **6. Use Insigths**

* After saving, leave the studio and click the **Insights** button to view interface or click in Ctrl+I.

***

#### Best Practices

* **Choose meaningful dimensions:** Avoid using too many dimensions at once—stay focused to avoid diluting the results. Use intuitive names to make insights easier to interpret.
* **Use relevant metrics:** Select numeric fields such as `sales`, `quantity`, or `profit`. Avoid metrics with many null values.
* **Always include a date field:** Date fields allow time-based analysis (e.g., monthly growth). Use fields like `sale_date` or `created_at`.
* **Set a default period:** In the *Default Period* field, choose a time frame like Monthly or Weekly to guide the analysis.
* **Use the growth threshold wisely:** Define a minimum percentage (%) to highlight only significant changes and ignore small fluctuations.
* **Avoid incomplete fields:** Make sure selected fields have clean and consistent data.
* **Schedule when appropriate:** Use the *Schedule* tab to automate insights generation (e.g., every Monday at 8 AM).
* **Add descriptions to fields:** Makes it easier for other users to understand and improves data governance.