Product speed has a large, measurable effect on conversion rates and user experience on your product. We did an analysis of OneSchema versus Flatfile’s performance across a variety of file sizes and performance categories, including upload, validation speed, and transform speed.
Product speed has a large, measurable effect on conversion rates and user experience on your product. The difference in conversion rate difference between a site that loads in 1 second is 3x higher than a site that loads in 5 seconds.
We did an analysis of OneSchema versus Flatfile’s performance across a variety of file sizes and performance categories, including upload, validation speed, and transform speed.
OneSchema vs. Flatfile Performance
Time for file to be uploaded, parsed, and intelligently map column headers for the user to confirm.
Note: Upload speed is dependent on the user’s network speed. All tests done in this article were performed using a network speed of 250Mbps.
Definition: Validation speed is the time it takes to go through the different validations for all mapped columns across the entire file. In this case, the tested validations were across two proper case fields (first name, last name), email, date, and address fields.
Definition: Transform speed is the time required to automatically fix all errors detected in the test file using OneSchema native transformations
Note: Transforms (e.g. features like one click autofix and find-and-replace) are not natively supported by Flatfile. Both OneSchema and Flatfile support transformation via webhook, but the speed is primarily dependent on your webhook’s response times.
We used CSV files of 1k rows, 10k rows, 100k rows, 1M rows, and 10M rows to test performance at different file sizes.
All files were uploaded with a total of 11 columns
After uploading, 5 columns were mapped and validated. This included two name columns (e.g. first name, last name), email, date, and address fields.
Tests for each file size and performance category were performed three times and then averaged to produce the final value.
Note: “Proper Case” validation is used for name columns, which is one of OneSchema’s slowest performing validators.
Engineering and product teams use OneSchema to easily build out best-in-class data import capabilities for their customers. Instead of spending months writing data validations and handing CSV parsing edge cases, focus your development resources on your core product development. With minimal engineering investment, improve your customer activation rates and launch better onboarding than your competitors.