Retail and eCommerce
Data Infrastructure and Analytics
A new skincare retailer was making stock decisions based on assumptions. We replaced that with a live, automated intelligence system.
Services
Data Centralisation
Competitor Intelligence
Inventory Automation
BI and Dashboards
Entering a competitive market without a data foundation is a significant disadvantage. This retailer had no centralised system, no competitor visibility, and inventory decisions made from spreadsheets. Dataverse built the entire data infrastructure from scratch and handed it over to a non-technical team to run independently.
Before this, stocking decisions were basically educated guesses. Within three months of having the data in front of us, sales were up fifteen percent. The team could see exactly what was working and where to focus.
15%
Sales increase within three months
40%
Reduction in manual analysis workload
10
High-performing categories identified for priority stocking
The Challenge
The retailer had entered the UK skincare market with no centralised data system. Understanding which products were gaining traction, which categories were worth prioritising, and what competitors were doing required slow, manual work across spreadsheets. Inventory accuracy suffered as a result, with replenishment decisions based on assumptions rather than evidence. The business needed a clear, reliable view of both the market and their own stock before scaling was possible.
No centralised product or inventory data, with everything managed across spreadsheets
No visibility of competitor product ranges, pricing, ratings, or category performance
Replenishment decisions made on assumptions, leading to stockouts and missed sales
Small non-technical team with no capacity to build or maintain data tooling
The solution
Dataverse built a complete data and insights ecosystem from the ground up. Python-based scraping pipelines using BeautifulSoup and Selenium extracted product data from four major competitors. All data was cleaned, standardised, and loaded into BigQuery, creating a unified dataset with consistent categories and measurements. Looker Studio dashboards were built with filters for category, brand, ratings, and reviews, giving the team real-time market visibility without any technical skill required. A custom Airtable environment was built to manage their own inventory, with every product assigned a unique identifier to track variations accurately. A Python script maintained data consistency and updated records automatically. API connections between Airtable and BigQuery kept inventory levels live, and low-stock thresholds triggered automated supplier alerts. The team was trained end-to-end to operate the system independently.
Competitor scraping pipelines across four major brands using BeautifulSoup and Selenium
Unified BigQuery dataset with standardised product categories, sizes, and measurements
Looker Studio dashboards providing live market trend visibility for a non-technical team
Custom Airtable inventory environment with unique product identifiers and automated consistency checks
Automated low-stock alerts to suppliers triggered via API connection between Airtable and BigQuery