Project Brief · Commercial in confidence

Python Developer for Ecommerce Business Attribution & Shopify Research Tool

Project Brief | Local Python research and automation tool for authorized ecommerce asset discovery and evidence-based attribution

Core objective

Discover potentially related Shopify stores, ecommerce sites, domains, and brand assets from incomplete authorized business information, then validate each relationship using public evidence.

Priority

Accuracy and explainability over volume. The tool must be comfortable returning No Reliable Match when evidence is insufficient.

01Project Overview

We are looking for an experienced Python developer to build a locally runnable research and automation tool for identifying ecommerce websites and Shopify stores associated with authorized businesses.

This is not a standard scraping project with a predefined list of URLs. The main challenge is discovering potential online business assets from incomplete business information and then determining whether the discovered assets are genuinely connected to the target business.

02Possible Case Inputs

Possible Case Inputs

Business or company name
Business address
Phone number or email address
Known website, domain, or privacy-policy URL
Brand or product information
Other publicly available business identifiers

Some cases may contain only a few of these fields.

What the Tool Should Discover

Shopify stores and ecommerce websites
Business and brand domains
Brand websites and related public online business assets

03Evidence Validation & Entity Resolution

Finding a candidate website is not enough. The system should determine whether sufficient public evidence connects that candidate to the original business. Evidence may include matching or related business names, addresses, phone numbers, emails or domains, contact information, privacy or terms pages, legal entity references, linked websites, social accounts, Shopify/storefront metadata, and other publicly visible signals.

The system should combine multiple signals rather than making a conclusion from a single weak indicator.

04Result Classification

Classification

Meaning

Strong Match

Multiple reliable signals support the relationship.

Possible Match

Some evidence exists, but it is not sufficient for confirmation.

Insufficient Evidence / No Reliable Match

Available public information does not reliably support the relationship.

Every result should include supporting source URLs and an explanation of why it received its classification. The goal is an auditable evidence trail, not a black-box confidence score.

05Expected Workflow

Case Input → Input Normalization → Candidate Discovery → Website / Shopify Detection → Evidence Extraction → Entity Resolution → Confidence Classification → JSON / CSV Output

06Technical Direction

The application should be written primarily in Python and run locally. Search APIs, scraping/extraction modules, Shopify detection, matching logic, and output generation should be modular so additional public data sources can be added later.

Relevant technologies may include: Python, Requests, BeautifulSoup, Selenium or Playwright where appropriate, search APIs, Pandas, structured web extraction, domain/website analysis, Shopify storefront data, fuzzy/entity matching, JSON, and CSV processing.

07Implementation & Validation

For the initial phase, representative authorized research cases will be provided. The developer should first establish and validate the research methodology against these cases, then implement the proven workflow as a reusable Python tool.

Accuracy is more important than quantity. The tool must be able to return "No Reliable Match" instead of making unsupported connections.

08Final Deliverables

Working locally runnable application
Modular research/source pipeline
Evidence-based matching and classification logic
JSON and/or CSV structured results
Source and evidence tracking

Error handling and logging
Configuration and API-key handling
Installation and operating documentation
Reliable live demonstration using agreed sample cases

09Authorized Research Scope

All research is limited to businesses and assets that are owned, controlled, or authorized for audit, using information that may lawfully be researched from public sources. The project does not involve unauthorized account access, bypassing access controls, private investigations, skip tracing, or obtaining non-public information.

10Developer Background

Please describe relevant experience with Python web research, Shopify/ecommerce scraping, entity resolution, search APIs, public-web data extraction, or evidence-based matching systems. Also briefly explain how you would approach candidate discovery and validation when only partial business information is available.