MediaAgility Builds a Machine Learning & Streaming Analytics Based, Real-time Store Dashboard for a Multinational Retail Brand

“The brand has optimized staff availability, inventory, and gained business-critical insights with the Streaming Analytics and Machine Learning based solution. The solution now handles 50K orders per day and 25K MPoS per day.”, confirms Asheesh Sharma, Partner at MediaAgility.

The Customer is a multinational clothing and lifestyle retail brand with hundreds of stores across two continents. They focus on highlighting the culture of their consumers’ countries with their creative apparels and accessories, both for men and women. They also house a wide range of innovative home products.

Pre-solution Days: No Visibility into Store Traffic Data & Performance

The Brand needed to derive insights from their sales, store traffic, conversions, and historical data to deliver customized consumer experiences and ensure consumer inclusion into their brand’s story. However, the data silos at the multiple stores posed these challenges:

  • Store managers had no visibility into intraday sales, traffic volumes, or conversions
  • There was no way to benchmark a day’s performance against business milestones

Due to lack of insights, store managers struggled to deliver seamless and memorable experiences to the consumers.

Analytics & Machine Learning to Derive Real-time Insights from Data

Certified Cloud Architects at MediaAgility, the retail brand’s digital innovation partner, designed and developed a solution to address their challenges. The approach was to –

  • Integrate data from data silos for strong data foundation
  • Generate real-time insights with streaming analytics

MediaAgility architects used the GCP stack to deliver the solution in less than 15 weeks –

  • Big Query
  • CloudSQL
  • Dataflow
  • Composer
  • Microstrategy
  • PubSub

The solution ingests real-time data from sales, store traffic, conversions, weather forecasts, holiday spikes, and historical, into the algorithm from the brand’s hundreds of stores. DataFlow provides real-time ETL, and data with different aggregation levels are fed into BigQuery and CloudSQL. Through streaming analytics on real-time data from their multiple stores, the solution provides insights into stores’ performances on a live dashboard simultaneously.

Business Benefits & Optimization

The Customer has achieved the following with the Machine Learning and Streaming Analytics-based solution –

  • Real-time aggregation and consolidation of data
  • Highly scalable and flexible data pipeline that grows with the brand
  • Visibility into intraday sales, transactions, and conversion data with business benchmarking
  • Faster reporting and decision making with customizable, real-time store dashboard
  • Optimization of staff and inventory, eventually leading to cost-savings