Catalogue collection
Names, attributes, prices and stock land in consistent columns.
Web scraping and data extraction
Data from multiple sources, delivered in one testable structure.
I collect products, prices, stock, contacts and statuses from websites, files and APIs. The output has clean fields, change history and explicit errors.
Discuss a project Input
websites and cataloguesExcel, CSV and price listsAPIs and messages
Logic
extractionnormalisationmatching and deduplication
Outcome
Google SheetsCSV or databaseAPI or dashboard
Manual work that can disappear
Supplier reconciliation
Multiple price lists are normalised and discrepancies are made visible.
Scheduled exports
Data refreshes on a schedule while source failures remain visible.
Project outcome
A reliable extraction workflow
- source and field map
- rate-aware collector
- validation and error log
- export format or API
- runbook and maintenance notes
Ten correct rows before ten thousand
- 01Map one workflow
Define the source, rules, exceptions and the outcome a user must see.
- 02Build a narrow test
Validate a small dataset or one user journey before a wider implementation.
- 03Launch and measure
Review errors, response time and actual value, then expand what works.
Data extraction FAQ
Can data be collected without site access?
Often yes when the information is public and available to a normal user. Each source is checked first.
What happens when a page changes?
The collector should flag missing or changed fields instead of silently writing bad data.
Can the output go directly to Google Sheets?
Yes. CSV, Excel, a database or an API are also possible.
Web scraping and data extraction