> ## Documentation Index
> Fetch the complete documentation index at: https://algolia.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Add CSV data to your extracted records
> Enrich your records with data from a CSV file.
If the data you want to include in your extracted records isn't present on the site you're crawling,
you can add that data from a CSV file.
For example, you may be crawling an ecommerce site, but details such as cost pricing and inventory levels aren't on the site.
## Prepare a spreadsheet in a crawler-compatible format
Your spreadsheet's format must be compatible with the crawler:
* It must contain a header row
* One of the header columns must be `url`
* The `url` column should contain all URLs to which you want to add external data.
* The remaining columns are for data that you want to add to the record for each URL.
For an example, see [this spreadsheet](https://docs.google.com/spreadsheets/d/1hkpL4yL5lmUajzpqs_MQqKnU3MBL3VQT1WLegNy76y8/edit?usp=sharing).
## Publish your spreadsheet online
Add your spreadsheet data to Google Sheets and [publish it online](https://support.google.com/docs/answer/183965).
* The spreadsheet's sharing settings(**Restrict access to the following**) must allow the Crawler to access it.
* The spreadsheet must be published as a CSV file.
Change the setting from **Web page** to **Comma-separated values (.csv)**.
* The crawler uses the latest data in your Google spreadsheet.
To prevent this, and just use the initially uploaded data, clear the
**Automatically republish when changes are made** setting.
After you click **Publish**, copy the URL.
You'll need this to create an external data source.
The CSV data doesn't have to be in Google Sheets but it must be available online.
## Link the published spreadsheet in your crawler configuration
To link your spreadsheet, create an external data source and add it to the `recordExtractor` function in your crawler configuration.
### Create an external data source
1. From the Crawler page, select the [External Data](https://dashboard.algolia.com/crawler/external-data) tab.
2. Click **Add External Data**
3. As **External Data type**, select **CSV**, add your CSV file's URL, and click **Create**.
4. To test the data source, click **Explore Data** and then **Refresh Data**.
It should extract the correct number of rows from your spreadsheet.
### Add external data to extracted records
1. Go to the [**Crawler**](https://dashboard.algolia.com/crawler/) page,
select your crawler, and click **Editor**.
2. Add the [`externalData`](/doc/tools/crawler/apis/configuration/external-data)
parameter to your crawler. For example:
```js JavaScript icon=code theme={"system"}
externalData: ["myCSV"];
```
3. Add the [`dataSources`](/doc/tools/crawler/apis/configuration/actions#param-record-extractor-data-sources)
parameter to the [`recordExtractor`](/doc/tools/crawler/apis/configuration/actions#param-record-extractor)
function and reference the columns from your CSV data source.
```js JavaScript icon=code highlight={4,5,7} theme={"system"}
recordExtractor: ({ url, dataSources }) => [
{
objectID: url.href,
pageviews: dataSources.myCSV.pageviews,
category: dataSources.myCSV.category,
// Convert the string to a boolean because CSV doesn't have boolean types
onsale: dataSources.myCSV.onsale === "true",
},
];
```
The crawler only keeps data from URLs that match your crawler configuration's [`startUrls`](/doc/tools/crawler/apis/configuration/start-urls),
[`extraUrls`](/doc/tools/crawler/apis/configuration/extra-urls), and
[`pathsToMatch`](/doc/tools/crawler/apis/configuration/actions#param-paths-to-match) parameters.
#### Test the data
1. In the [**URL Tester**](/doc/tools/crawler/getting-started/monitoring#url-tester),
enter the URL of a page with CSV data attached to it (one of those you added to your spreadsheet).
2. Click **Run Test**.
3. Confirm that the extracted records contain the data from your CSV.
If this doesn't work as expected,
check that the URL in your spreadsheet matches the site URL.