Scrape Retail Product Reviews
Budget: $30 – $250 USD
I need a clean, well-structured dataset of customer reviews pulled from Home Depot, Lowe’s, Amazon, SiteOne, and Ferguson. The file I’ve uploaded spells out every detail and includes direct product links and sample formats—please follow that guide closely.
Scope
• Cover both Stormwater Management and Landscape Irrigation items, but only models that are fully plastic.
• Stormwater Management lines: Catch Basin, Grates, and Channel Drain.
• Landscape Irrigation lines: Access/Valve Boxes, Drippers, and Irrigation or Electric Hose Timers.
• Check every relevant listing from the manufacturers named in the brief (NDS, RELN, Polylok, U.S. Trench Drain, Vevor, ADS for stormwater; NDS, Raindrip, Rain Bird, Orbit, DIG, Mister Landscaper, Hunter for irrigation).
Deliverables
• A single CSV or Excel file based on the example file that was uploaded - it will have one tab per retailer.
• A brief read-me explaining your extraction method (Python + BeautifulSoup/Scrapy/Selenium, etc.) and any site-specific caveats.
• The reproducible scripts or notebooks so the scrape can be rerun later.
Acceptance Criteria
The dataset must reflect all live reviews on the listed retailers for every qualifying product at the time of scrape, deduplicated and UTF-8 encoded. Spot-checks against the live pages should show 100 % accuracy on ratings and text.
Timeline is flexible over the next few days, but let me know if certain sites require additional anti-bot steps so we can adjust.
Scope
• Cover both Stormwater Management and Landscape Irrigation items, but only models that are fully plastic.
• Stormwater Management lines: Catch Basin, Grates, and Channel Drain.
• Landscape Irrigation lines: Access/Valve Boxes, Drippers, and Irrigation or Electric Hose Timers.
• Check every relevant listing from the manufacturers named in the brief (NDS, RELN, Polylok, U.S. Trench Drain, Vevor, ADS for stormwater; NDS, Raindrip, Rain Bird, Orbit, DIG, Mister Landscaper, Hunter for irrigation).
Deliverables
• A single CSV or Excel file based on the example file that was uploaded - it will have one tab per retailer.
• A brief read-me explaining your extraction method (Python + BeautifulSoup/Scrapy/Selenium, etc.) and any site-specific caveats.
• The reproducible scripts or notebooks so the scrape can be rerun later.
Acceptance Criteria
The dataset must reflect all live reviews on the listed retailers for every qualifying product at the time of scrape, deduplicated and UTF-8 encoded. Spot-checks against the live pages should show 100 % accuracy on ratings and text.
Timeline is flexible over the next few days, but let me know if certain sites require additional anti-bot steps so we can adjust.