# Charts and widgets With CARTO for Developers you can easily build customizable and interactive data widgets, using live data from your cloud data warehouse, on top of your [CARTO + deck.gl](https://github.com/CartoDB/gitbook-documentation/blob/master/carto-for-developers/key-concepts/carto-for-deck.gl) map.

An example using a Formula widget and a Categories widget

## What is a widget? Widgets in CARTO are a series of flexible **data models** from the same data sources you can use in your maps. For example, the *Formula widget* represents a single numerical statistic, such as an average or a count out of the column you define. And the *Categories widget* allows you to group and aggregate based on columns of your choice. You can then turn those data models into **interactive charts, filters and other HTML elements**. ## How do I create widgets? Adding widgets is easy. You just need to define your [data sources](https://docs.carto.com/carto-for-developers/key-concepts/data-sources) from `carto-api-client` just like you do in the CARTO module in deck.gl, and then use any of their `widgetSource` data models, such as `yourSource.widgetSource.getCategories()`:

import {VectorTableSource} from 'carto-api-client';

// define auth parameters in cartoConfig
  
const dataSource = await vectorTableSource({
  ...cartoConfig,
  tableName: 'carto-demo-data.demo_tables.osm_pois_usa'
})

const categories = await dataSource.widgetSource.getCategories({
  column: 'group_name',
  operation: 'count'
}) 

// categories will return the corresponding aggregation by the column group_name,
// you can then plug categories into any charting library or HTML element

A widget is a purposefully ample concept because it can be transformed into pretty much anything. A widget in CARTO can turn into * **Custom HTML data representations:** such as scorecards, tables or natural text. * **Charts and other visualizations:** you can use your favorite JS charting library to build bar charts, pie charts, radar charts and other charts. * **Inputs and other interactive elements:** some data models work great to build interactive inputs (such as range sliders) or you can easily build cross-filtering when users click in your charts.

## Get started * Complete [technical reference for CARTO Widgets](https://docs.carto.com/carto-for-developers/reference/carto-widgets-reference) * We have prepared a set of [examples showcasing charts and widgets](https://docs.carto.com/carto-for-developers/examples) ## Community Our community in Slack is a great place to ask questions, share your work with other CARTO users, and get help from CARTO experts. **Join here** :arrow\_down: {% embed url="" fullWidth="false" %}