Bulk runs can be started from a robot's Run Task tab, or in a robot's data Table by using the Import CSV option.
In both cases, you can upload a CSV containing up to 500,000 different sets of Input Parameters and immediately create a task for each one of them.
A bulk run allows you to extract data from multiple URLs in one go by uploading a CSV file containing up to 500,000 different input parameters. This feature helps you build comprehensive datasets efficiently across many pages from the same website.
Executing a bulk run via the Run Task tab
You can execute a bulk run directly from your robot's Run Task tab.
Click the Bulk Run Tasks button.
Download the Sample CSV Input if needed to format your CSV correctly.
Upload the prepared CSV file to run the robot using the included input parameters.
Executing a bulk run via the Table import CSV option
You can execute a bulk run from Tables via the import CSV option.
Click the Import CSV button at the bottom of your table.
Download the Sample CSV Input if needed to format your CSV correctly.
Select the Click to upload area, then choose your desired CSV.
I've started my bulk run - what happens next?
Once you've executed your bulk run, the tasks will be processed in a queue. You will get the fully extracted dataset once the bulk run has finished.
If you encounter failures during bulk run execution, it may indicate insufficient credits in your Browse AI account. Verify and ensure you have adequate credits to complete the task before starting. Visit your robot's History tab to monitor the progress and adjust configurations as necessary.
My bulk run has completed - where do I see the results?
You can have a holistic view on these results, as well as export the very same results, via our Tables.
See:
Troubleshooting Bulk Run Issues: One common issue is running out of processing credits, which can halt tasks. Ensure your account has sufficient credits and monitor tasks efficiently using logs.
What are example use cases where a bulk run is useful?
Bulk runs are great for when you want to extract similarly structured data across multiple web pages across the same site.
For example, let's say I wanted to create a company database using data from LinkedIn. Bulk monitoring would allow me to upload a CSV containing links of up to 500,000 company pages on LinkedIn, and then extract that data from all of those pages into one dataset.
See:
Best Practices for Successful Bulk Task Execution
Plan Credit Usage: Evaluate the estimated number of tasks and ensure sufficient credits are available before initiating the process.
Standardize URLs: Verify that all URLs match a consistent structure for compatibility with the same robot.
Test Runs: Conduct small-scale test runs before applying a bulk run task to the entire dataset.
Monitor and Adjust: Regularly check task logs to address any errors and fine-tune configurations for optimal results.