Browse categories
Explore
Fiverr Pro
English
$
USD
I will do web scraping into a clean CSV dataset
About this Gig
Turn matching public website tables into one clean CSV, with a source URL and row number for every record.
Send 1-100 explicit public HTTPS URLs from one origin (same host), the same table number and up to 10 exact column headers. All pages need the same static table layout. Each source table needs at least one data row. This $30 package covers 1,000 total data rows, 2-day delivery and 1 revision.
Receive the CSV, captured source bundle, offline replay script, report, README and AI-use disclosure. Independent checks reconcile every requested page; rows are never silently dropped or invented.
Limits: 2 MB per page, 10 MB total HTML, 20 MB source bundle, 10 MB CSV and 40 MiB package. Robots permission and required waits up to 5 minutes per request are enforced. Blocked, oversized, unsupported or incomplete batches are held for review; no partial delivery.
No login, JavaScript rendering, URL discovery, merged/nested tables or ongoing updates. Source tables must have at most 50 columns. Revisions change selected columns using frozen captures; new URLs, tables or refreshed data need new scope. Work is AI-assisted; no calls or account access. Import CSV columns as text.
Technology:
Python
Information type:
Websites
Technique:
Automated
FAQ
What will I receive?
One CSV with your selected columns plus source_url and source_row, the captured source bundle, an offline replay script, report, README and AI-use disclosure. The $30 package covers up to 100 URLs, 1,000 total rows and 10 selected columns, with 2-day delivery and 1 revision.
Which websites are supported?
Public HTTPS pages on one origin, with the same static HTML table layout and a header plus at least one data row per page. No login, JavaScript, discovery or merged/nested tables. Automated access must be permitted; robots waits over 5 minutes per request are outside scope.
What are the file and data limits?
Up to 100 URLs, 1,000 aggregate data rows and 10 selected columns. Source tables may have up to 50 columns. Limits: 2 MB per page, 10 MB total HTML, 20 MB source bundle, 10 MB CSV and 40 MiB package. Every page must pass; blocked, oversized or incomplete batches are held.
What does the revision cover?
One revision to the selected column list, within the 10-column limit, using the same frozen captures. New URLs, another table, refreshed source data and ongoing updates need new scope. The included script replays the captured source offline; it does not crawl or monitor websites.
How are the data checked?
Independent checks reconcile every page and selected cell, URL order, row counts, provenance and file hashes. The delivered script must reproduce the CSV offline. Whitespace is normalized; blanks stay blank. No values are invented. Import every CSV column as text to avoid formulas.
