McFadyen Digital · Case Studies

Real-world AI in Commerce

205 documented AI implementations across commerce — filterable by industry, value chain, technology, and outcome. Every entry is sourced and quantified where the data exists.

205
Implementations
12
Industries
16
AI technologies
Documented at companies like

Showing 117 of 17 case studies

Filters
4 min read
Clothing and Clothing Accessories Retailers2025
Global lifestyle and beauty brand

A global lifestyle player deploys a generative AI shopping assistant and lifts conversion rates by up to 20%

A global lifestyle and beauty brand deployed a gen-AI-powered LLM shopping assistant trained on product data and consumer preferences — achieving up to a 20% lift in conversion rates, reported in a McKinsey analysis of gen AI in the beauty sector.

up to 20%Conversion Rate Lift
Generative AIB2C
Read →
5 min read
Computer Systems Design and Related Services2025
Amazon Web Services (Virtual Engineering Workbench)

AWS-powered Virtual Engineering Workbench cuts automotive test-case creation time by up to 80% using Amazon Bedrock

An AWS Professional Services team integrated Amazon Bedrock into the Virtual Engineering Workbench (VEW) for automotive software testing, reducing test-case creation time by up to 80% while maintaining accuracy through a human-in-the-loop approach. The solution was implemented in production in four weeks.

Up to 80%Test Creation Time Saved
4 weeksProduction Deployment
Generative AIB2B
Read →
4 min read
Computer Systems Design and Related Services2025
Microsoft

Microsoft automates end-to-end bug triage with two AI agents in Azure DevOps, reducing time from identification to resolution

Microsoft built and published an Auto Triage AI solution using Copilot Studio and Azure DevOps, using two autonomous agents to extract bug details from customer emails, generate reproduction steps, create bug records, and post follow-up updates automatically.

2 autonomousAgents
Generative AIB2B
Read →
5 min read
Grocery and Convenience Retailers2024
ASDA

ASDA scales mobile test coverage by 50% and cuts manual testing effort by 70% with Perfecto AI

UK supermarket giant ASDA adopted Perfecto AI's cloud-based testing platform to automate mobile and web testing across its ecommerce site, mobile app, Scan & Go, and Rewards apps — achieving a 50% increase in test coverage, 70% reduction in manual testing effort, and a 4x expansion in device and OS coverage.

50%Test Coverage Increase
70%Manual Testing Reduced
Generative AIB2C
Read →
5 min read
Clothing and Clothing Accessories Retailers2024
Bestseller

Bestseller builds 110 automated test cases in 3 months with Leapwork, democratizing QA across fashion brands

Denmark-based fashion group Bestseller implemented Leapwork's low-code test automation platform, enabling functional business users to create and maintain test flows across Dynamics 365, POS, and e-commerce systems — without relying exclusively on technical staff.

110 in 3 monthsTest Cases Built
120 subflowsReusable Components
Machine Learning (classification)B2C
Read →
5 min read
Credit Intermediation and Related Activities2024
Deutsche Bank

Deutsche Bank cuts test data provisioning time 80% with synthetic data for ESG onboarding

Deutsche Bank's Credit Risk Technology team adopted the Synthesized platform to generate high-fidelity, privacy-safe test data for ESG onboarding workflows, cutting data access delays dramatically and embedding synthetic data into CI/CD pipelines.

80% reductionData Provisioning Time
Generative AIB2B
Read →
5 min read
Computer Systems Design and Related Services2024
NVIDIA

NVIDIA's Hephaestus framework saves pilot teams up to 10 weeks of development time with AI-generated tests

NVIDIA's DriveOS team built Hephaestus (HEPH), an internal generative AI framework that automates test-case creation from software architecture and interface control documents, saving pilot teams up to 10 weeks of development time per engagement.

Up to 10 weeks (per pilot team)Dev Time Saved
Generative AIB2B
Read →
4 min read
Grocery and Convenience Retailers2023
Auchan

Auchan unifies web and mobile testing with Katalon to accelerate e-commerce DevOps delivery

French multinational retailer Auchan adopted the Katalon platform including Runtime Engine, TestCloud, and TestOps to automate web and mobile testing across browsers and devices, reducing manual regression work and improving release reliability for its e-commerce and order management systems.

Web + Mobile unifiedTesting Scope
Machine Learning (classification)B2C
Read →
5 min read
General Merchandise Retailers2023
Walmart

Walmart deploys AI/ML forecasting across 4,700 stores to prevent holiday stockouts and supply chain disruptions

Walmart's centralized AI/ML inventory platform analyzes demand at the individual item-store-day level, enabling proactive replenishment, geographic redistribution of inventory, and anomaly 'forgetting' to avoid over-correcting on one-time events.

4700 locationsStores Connected
up to 90% improvement (unverified)Inventory Accuracy
ForecastingB2C
Read →
5 min read
Computer Systems Design and Related Services2022
eBay

eBay scales 8 petabytes of production data into 1 GB de-identified test subsets with Tonic.ai, unblocking automated testing

eBay's 8-PB distributed data ecosystem made staging unreliable and automated testing expensive. Using Tonic.ai's subsetting and de-identification platform, eBay provisioned privacy-safe 1 GB subsets across ten critical domains — shortening development cycles and increasing automation pass rates.

8 PB petabytesProduction Data Managed
1 GB per domainTest Subset Size
Generative AIMarketplace
Read →
5 min read
Computer Systems Design and Related Services2022
Microsoft

Microsoft's flaky test management system spans 100 product teams, flags 49,000 flaky tests, and prevents 160,000 unnecessary session failures

Microsoft built a company-wide flaky test management service — embedded in CloudBuild and CloudTest — that supports over 100 product teams, has identified approximately 49,000 flaky tests, and has prevented more than 160,000 test sessions from failing unnecessarily, dramatically improving developer productivity across its global engineering organization.

49,000 testsFlaky Tests Identified
160,000 sessionsSessions Saved
Machine Learning (classification)B2B
Read →
5 min read
Broadcasting and Content Providers2022
Netflix

Netflix's Pensive system auto-diagnoses and remediates failed big-data jobs across hundreds of thousands of daily workflows

Netflix's AI-driven Pensive platform uses a rules engine augmented by ML clustering to classify and automatically fix failed batch and streaming data jobs — reducing manual operational burden across the world's largest cloud-based streaming data platform.

hundreds of thousands workflows/dayDaily Workflows
Batch + Streaming workload typesPlatform Coverage
Machine Learning (classification)B2C
Read →
6 min read
Computer Systems Design and Related Services2022
Slack

Slack's Project Cornflake cuts flaky test job failures from 56.8% to 3.9% and saves 553 hours of triage time

Slack's Mobile Developer Experience team built an automated flaky test detection and suppression system — Project Cornflake — that reduced test job failure rates from 56.76% to 3.85% in under a year, saving 553 hours of manual triage time across Android and iOS mobile codebases with 27,000+ automated tests.

56.8% → 3.9% reductionTest Job Failure Rate
553 hoursTriage Time Saved
Machine Learning (classification)B2B
Read →
4 min read
Computer Systems Design and Related Services2021
eBay

eBay scales 8 PB data ecosystem to 1 GB test subsets with Tonic.ai, accelerating release velocity

eBay used Tonic.ai's synthetic data and subsetting platform to give developers on-demand access to production-like test data, eliminating staging unreliability and fueling faster automated testing cycles.

8 PB→ 1 GB subsetsData Scaled
Higher pass rateAutomated Testing
Machine Learning (classification)Marketplace
Read →
5 min read
Computer Systems Design and Related Services2020
GitHub

GitHub reduces flaky builds by 18x — from 9% of commits to less than 0.5% — with automated detection and quarantine

GitHub built an internal flaky test management system for its Ruby on Rails monolith that reduced the percentage of commits with at least one flaky build from approximately 9% (1 in 11 commits) to less than 0.5% (1 in 200), an 18x improvement, by automatically detecting, containing, and delegating flaky failures to the engineers who introduced them.

18x improvementFlaky Build Reduction
9% → <0.5%Commits with Flaky Builds
Machine Learning (classification)B2B
Read →
5 min read
Computer Systems Design and Related Services2018
Meta Platforms

Meta's predictive test selection captures 99.9% of regressions while running only one-third of tests, doubling infrastructure efficiency

Meta (formerly Facebook) deployed a machine-learning predictive test selection system that catches more than 99.9% of regressions before they reach the trunk, while running just a third of all dependent tests — halving testing infrastructure costs.

>99.9%Regressions Caught
~33% (of dependent)Tests Run
ForecastingB2C
Read →
4 min read
Clothing and Clothing Accessories Retailers2016
Gymshark

Gymshark's 8-hour Black Friday crash cost £100,000 and forced a full ecommerce replatform to Shopify Plus

After its Adobe Commerce site collapsed under Black Friday traffic in 2015, fitness apparel brand Gymshark migrated to Shopify Plus — reaching £41 million in annual sales by 2017 and processing thousands of orders per minute without an outage.

8 hoursDowntime
£100,000 GBPEstimated Loss
Optimization / Operations ResearchD2C
Read →