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
Walmart launches generative AI search on Azure OpenAI Service, contributing to 22% global ecommerce growth in Q1 FY2025
Walmart built a generative AI-powered search engine using Azure OpenAI Service and proprietary retail models, enabling shoppers to submit natural-language, occasion-based queries and receive curated product bundles. The company reported 22% global ecommerce growth in the quarter when the capability went live.
Walmart launches generative AI search on Azure OpenAI Service, contributing to 22% global ecommerce growth in Q1 FY2025
Walmart built a generative AI-powered search engine using Azure OpenAI Service and proprietary retail models, enabling shoppers to submit natural-language, occasion-based queries and receive curated product bundles. The company reported 22% global ecommerce growth in the quarter when the capability went live.
Carson Group Achieves 96% Lead Conversion Accuracy with AWS-Powered ML Scoring
Financial advisory firm Carson Group, managing $57B in assets, deployed a machine learning lead scoring model built by Provectus in five weeks — reaching 96% prediction accuracy and 20× faster lead evaluation.
ihomer cuts duplicate code 20% and achieves zero critical AI-code vulnerabilities with Codacy Guardrails
Netherlands-based EV charging software provider ihomer deployed Codacy to govern AI-generated code, achieving a 20% reduction in duplicate code across key repositories, zero outstanding critical vulnerabilities from AI-assisted code, and 50%+ developer adoption of AI guardrails in daily workflow.
Stripe's AI coding claim — '40% of routine code snippets' attributed to Stripe but not verified in primary sources
The book states that Stripe found AI tools could generate up to 40% of routine code snippets; no Stripe primary source supports this figure. Stripe's verified AI coding metric is 1,300+ AI-authored pull requests merged per week via its Minions autonomous coding agents (2026).
Ulta Beauty Deploys AI Recommendation Platform to Personalize Product Combinations for 46M+ Members
US beauty retail leader Ulta Beauty built Quazi, a proprietary AI personalization platform, to recommend individualized product combinations across marketing, social, and sales channels — leveraging insights from its 46M+ member loyalty base.
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.
Amazon AWS cuts video accessibility cost with generative AI audio description system
Using Amazon Nova, Rekognition, and Polly, AWS engineers demonstrated an automated pipeline that generates narrated audio descriptions for video content—potentially replacing a process estimated at $25 per minute when done manually. Still an early experiment, not a production deployment.
Anonymous manufacturer cuts procurement costs 40% using AI-powered supplier negotiation and contract intelligence
An unnamed manufacturer working with eMoldino deployed NLP-based contract intelligence, predictive analytics, and automated negotiation bots, cutting procurement costs by 40% and reducing contract review time by 60%.
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.
AWS Virtual Engineering Workbench cuts automotive test case creation time by up to 80% using generative AI
AWS built a generative AI extension of its Virtual Engineering Workbench that automates test case creation from automotive software requirements — reducing test case creation time by up to 80% while maintaining a human-in-the-loop review step. In a real-world deployment with Schaeffler, the system cut an experienced test engineer's preparation time from 820 hours to 265 hours for 837 requirements.
Broadcom resolves 88% of IT support requests automatically with Moveworks AI, down from days to under 60 seconds
Broadcom deployed Moveworks as a conversational AI assistant for internal IT support, achieving 88% autonomous resolution of IT requests and reducing average resolution time from days to under 60 seconds—with more than 75,000 total issues resolved.
Coca-Cola's Project Fizzion generates hundreds of localized campaign variations across 130+ languages using AI-encoded brand guidelines
Built with Adobe Firefly generative AI, Fizzion encodes Coca-Cola's brand rules into machine-readable 'StyleIDs' that allow creative teams to produce localized content up to 10x faster—while pilot programs in AI-driven localization claimed 7–8% sales growth and 20% higher engagement.
Deutsche Telekom upskills 8,000 agents with AI coaching engine, lifting first-time resolution 10% and NPS by 14 points
Partnering with McKinsey's QuantumBlack, Deutsche Telekom built a hyper-personalized AI capability engine that reduced transferred calls by 2% and increased first-call resolution by 10%, while boosting customer satisfaction by 14 NPS points.
DSV deploys Locus Robotics RaaS to scale warehouse throughput for a health and beauty client during 200–300% seasonal demand peaks
Logistics provider DSV implemented Locus Robotics' robots-as-a-service (RaaS) platform for a health and beauty retailer facing twice-yearly demand surges of 200–300%, enabling rapid scale-up and scale-down without permanent capital investment in equipment.
Global gaming hardware leader cuts localization costs 57% and translates 140% more content with Milengo's AI workflow
A $1B+ gaming hardware manufacturer operating in 75 countries replaced a human-heavy translation workflow with Milengo's AI-enhanced machine translation engine and custom termbases in 15+ languages, achieving 57% cost savings while translating 140% more content within the same budget.
Halliburton cuts tool retrieval time 22% with Navigine BLE/UWB real-time tracking across production sites
Energy services giant Halliburton deployed Navigine's RTLS platform with BLE AoA and UWB tracking across indoor arenas and outdoor production pads, achieving 22% faster tool retrieval, 18% reduction in equipment idle time, and 30% improvement in training-to-field execution consistency.
HCLTech client achieves 25% faster software development and 60% productivity gain with Google Gemini Code Assist
HCLTech deployed Google Gemini Code Assist for an unnamed client (likely Mattel, based on source asset metadata), delivering 25% faster development from scratch, 60% improvement in unit test coverage productivity, and 80% time savings in manual scripting — without adding headcount.
Indiana Pacers achieve 1.14% caption error rate with Azure AI live arena captioning
Pacers Sports & Entertainment built the NBA's first live in-arena captioning service using Azure AI Foundry, cutting speech-to-text error rates by 87% and reducing model training time from five days to two minutes.
Ma'aden saves 2,200 hours monthly in first phase of Microsoft 365 Copilot deployment
Saudi mining giant Ma'aden deployed Microsoft 365 Copilot, Copilot Studio, and Azure OpenAI Service, saving 2,200 hours per month across 36,600+ interactions in the first phase of rollout.
Macy's robot-equipped fulfillment center ships orders in under one day, roughly halving previous fulfillment time
Macy's new $640 million, 2.5 million-square-foot automated fulfillment center in North Carolina uses robotics and AI to complete orders in less than one day on average — compared to one to one and a half days at conventional warehouses — and more than doubled productivity at its first two automated sites.
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.
Microsoft's AI-powered continuous improvement framework boosts network performance metric 40% with AI agent
Microsoft embedded generative AI into its enterprise continuous improvement methodology, with an AI agent helping responsible individuals resolve network outages faster and delivering a 40% boost to a key network performance metric.
Microsoft's internal AI code review assistant covers 90% of 600,000 pull requests per month, cutting PR completion time 10–20%
What began as an internal experiment at Microsoft has scaled to support over 90% of the company's pull requests — more than 600,000 per month — with early data showing a 10–20% median improvement in PR completion time across 5,000 onboarded repositories.
Morrisons reaches 50,000 daily searches with Gemini AI in-app product finder built on Google Cloud
UK supermarket chain Morrisons launched an in-app Product Finder powered by Gemini AI models on Google Cloud — using BigQuery and Cloud Run — to help shoppers instantly locate items by aisle and shelf position. The tool saw 50,000 daily uses during the Easter 2025 peak period, reducing shopper frustration during seasonal store rearrangements.
Newell Brands produces creative campaigns 5x faster with Adobe Firefly Custom Models
Elmer's slime campaign assets that once took months to plan and shoot are now produced in days; Yankee Candle's European team generates dozens of design variations in seconds—all from a single AI-powered content supply chain.
NTNU empirical study finds AI-assisted backlog grooming achieves 100% precision and 45% faster duplicate detection than manual review
In a controlled study of 5 participants and 30 Jira issues, a generative AI tool operated as a human-in-the-loop assistant achieved 100% precision on duplicate detection while reducing time per duplicate found by 45% versus manual grooming.
P&G, Harvard, and Wharton study finds AI-assisted teams work 12% faster and innovate more broadly
A live hackathon with 776 P&G employees conducted with Harvard Business School and Wharton found teams using generative AI (GPT-4 via Azure) completed tasks 12% faster and professionals from different backgrounds developed more balanced solutions.
Perry Ellis lifts conversion 25% in two weeks with AI review summaries
Apparel brand Perry Ellis saw a 25% conversion-rate increase (1.1% to 1.4%) and a 12% rise in revenue per session within two weeks of piloting Yotpo's AI Review Summary, then rolled it out brand-wide. Amazon and Tripadvisor deploy similar AI review summarization at scale.
Sayvant saves 50,000 emergency clinician hours and cuts charting time 85% with Azure OpenAI
Built on Microsoft Azure OpenAI Service, Sayvant's ambient documentation platform reduces ER charting from 10 minutes to under 90 seconds per patient across 30+ languages, with a 40% drop in discharge delays at participating sites.
Uber deploys AI agents across customer support to surface context, automate investigations, and deliver empathetic resolutions
Uber's AI co-pilot for front-line agents provides conversational summaries, automates routine investigations, and translates complex policies into actionable resolutions, freeing staff for higher-value interactions.
Unnamed global consumer brand automates 90% of warranty claims and delivers 220% ROI in 12 months
A modular AI agent combining UiPath RPA and OpenAI LLMs auto-validates warranty claims in near real-time, saving $250,000+ annually and resolving claims up to 70% faster—while routing only high-value cases to human reviewers.
Walmart's Scintilla insight platform collects 3.4 million customer responses in 2024; subscribing suppliers grow omni sales 15% faster than non-subscribers
Renamed from Luminate to Scintilla in 2025, Walmart's supplier insight service gathered 3.4 million verified responses from its Customer Spark Community panel in 2024 — and suppliers who subscribe outpaced non-subscribers with 15% higher omni sales growth.
Zalando Transforms Thousands of Postmortems into Strategic Infrastructure Intelligence Using a Multi-Stage LLM Pipeline
Zalando's SRE team built a multi-stage AI pipeline — now powered by Claude Sonnet 4 on AWS Bedrock — that analyzes its entire postmortem corpus to surface recurring datastore failure patterns, cutting analysis time from days to hours while requiring ongoing human curation to address a persistent ~10% attribution error rate.
Access Holdings cuts software development tasks from eight hours to two with Microsoft 365 Copilot
Nigeria-based financial group Access Holdings deployed Microsoft 365 Copilot across its software development and operations teams, reducing an eight-hour coding task to two to three hours and compressing a chatbot build from two to three months to ten days.
AI-Driven Ticket Routing Delivers $3.70 Return per Dollar Invested, IDC Reports
A 2024 IDC study finds 63% of enterprises are adopting AI-driven automation for ticket handling; generative AI investments yield an average 3.7x return, driven by faster classification, reduced manual triage, and improved service reliability.
Amazon internal tests find AI-generated advertising imagery produces average 5% sales lift for participating brands
Internal Amazon testing of its AI Image Generator tool between October 2023 and June 2024 found that brands using AI-generated lifestyle imagery in advertising campaigns saw an average of nearly 5% more sales per advertiser — with click-through rates up to 40% higher for lifestyle versus plain product images.
Amazon removes Just Walk Out from Amazon Fresh grocery stores, shifts to Dash Cart as smart cart shoppers spend 10% more
Amazon's Just Walk Out technology — originally deployed in Amazon Fresh large-format grocery stores — was replaced in 2024 with Amazon Dash Cart smart shopping carts after customer research showed large-store shoppers preferred real-time spending visibility. Dash Cart shoppers at Amazon Fresh spend 10% more than non-Dash Cart shoppers, with 98% customer satisfaction and over 80% repeat usage rates. Just Walk Out continues to expand in small-format third-party locations.
Amazon uses A/B testing and causal ML to validate AI-enhanced catalog data improves customer experience
Amazon's Catalog team applies generative AI to enrich product data at scale and uses causal random forests and Bayesian structural time series to rigorously measure whether enrichments improve customer decisions and conversion.
Amazon uses AI to analyze competitor pricing and demand trends, adjusting ~2.5 million prices per day to protect margins while maintaining low-price perception
Amazon's AI-powered dynamic pricing system adjusts product prices approximately 2.5 million times per day — analyzing competitor pricing, demand trends, and customer behavior in real time — enabling the company to maintain its low-price brand perception while selectively improving margins on less price-sensitive products.
Artefact's GenAI Persona Generator produces accurate customer personas in 20 seconds — winning first place at e-Commerce Berlin Awards 2024
Germany-based marketing technology firm Artefact built the GenAI Persona Generator, which uses real customer data to create precise marketing personas in about 20 seconds — work that previously took several days of workshop preparation. The tool won first place in the Innovation category at the e-Commerce Berlin Awards 2024.
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.
Autodesk reduces incidents by 69% and cuts MTTR by 85% with BigPanda AI correlation
Autodesk, facing 100,000 monthly alerts across 25 monitoring tools, deployed BigPanda's AIOps platform and reduced incidents by 69% while improving MTTR by 85%—reclaiming engineering time for feature development.
Barbeques Galore scales from 1 to 70+ B2B dealers in 12 weeks after launching self-service portal on BigCommerce
Australia's leading grill retailer replaced a legacy B2B ordering system with BigCommerce B2B Edition in just 12 weeks, expanding its dealer network from one to more than 70 across Australia and New Zealand while enabling self-service quoting, invoicing, and ordering for trade buyers.
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.
Celonis cuts translation turnaround time 70% and halves time-to-market with AI-powered localization for its global training platform
Process mining software leader Celonis deployed Transifex's AI-powered localization hub for its Celonis Academy training platform, reducing translation cycles from ten days to two days and achieving 5x more monthly course completions across five languages.
DailyPay saves agents 40–60 seconds per call and $2M+ in costs with AI call summarization and Auto QA
By deploying Observe.AI's Conversation Intelligence Platform, Auto QA, and Summarization AI, DailyPay cut after-call work, improved CSAT by 22%, lifted service quality scores by 8.5%, and realized more than $2 million in operational savings.
Databricks Deflects 73% of Support Tickets with Agentic AI Assistant R2DB, Saving $1.5M Annually
After expanding its AI assistant R2DB beyond a narrow initial scope, Databricks achieved 73% ticket deflection across IT and HR support for 8,000+ employees, cutting hiring costs by $1.5 million and doubling employee IT satisfaction scores.
Datadog's Bits AI Drafts Postmortems from Incident Metadata and Slack Threads, Cutting Engineer Documentation Load
Datadog built an LLM-powered postmortem assistant inside its Bits AI product that combines structured incident metadata with unstructured Slack discussion, producing draft reports that give engineers a structured starting point rather than a blank page.
Demandbase Qualifies $3.5 Million in Pipeline in One Quarter Using G2 Buyer Intent Integration
By routing G2 buyer intent signals through its own ABM platform, Demandbase identified in-market mid-market accounts, surfaced 22 competitive and direct opportunities in a single quarter, and qualified $3.5 million in pipeline — demonstrating how intent data closes the gap between signal and revenue.
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.