A scraping, enrichment, deduplication, and Google Sheets pipeline for a 149,000-member UK business directory.
The client needed useful, structured business records from a large directory without manually collecting, cleaning, and reconciling entries.
Owned discovery, pipeline design, data extraction, normalization, enrichment, duplicate handling, delivery structure, and client handover.
Large directories contain inconsistent fields and duplicates. Delivery had to remain reviewable by a nontechnical team.
Client identity is confidential.
Targets, selectors, enrichment vendors, and record samples are withheld.
No unverified throughput or accuracy figure is claimed.
The workflow separated collection, normalization, enrichment, identity matching, and spreadsheet sync so each stage could be inspected and rerun independently.
Stage-separated processing
Duplicate detection before delivery
Structured review output
Private selectors and target-specific logic withheld
The delivered system produced a usable business-data pipeline and led to a joint-venture proposal, as recorded in the supplied resume.
Tayyab can translate difficult web data into a client-operable business workflow.
A clean delivery schema matters as much as collection.
Enrichment should remain separate from source-of-truth fields.
Sanitized architecture can still show engineering ownership.