All posts
Case Studies

What Ten Enterprise Modernizations Have in Common: Home Depot, Lowe's, Capital One, Netflix, Amazon and More

By

October 25, 2026

8 min read

Ten public modernization stories, side by side: where each started, how long it took, how much was rebuilt rather than moved, and what changed. Six patterns repeat in almost all of them, and each one applies to a company far smaller than these.

Case Study
Modernization
Cloud Migration
Strategy
Legacy Modernization

The ten, side by side

Company

Started with

Span

Approach

Reported outcome

Netflix

Customer-facing services after a 2008 database outage

2008–2016

Rebuilt as microservices; no lift-and-shift

Near four nines; cost per stream a fraction of the data center

Capital One

Applications across eight data centers

About 8 years, done 2020

80% of ~2,000 apps rebuilt cloud native

Dev environments in minutes; releases many times a day

Amazon consumer business

~7,500 Oracle databases, 75 PB

Done October 2019

Each workload to a purpose-built database

Costs down 60%+, latency down 40%, admin down 70%

The Home Depot

The enterprise data warehouse, then SAP

From 2016

Warehouse moved and retired; SAP moved

450 TB warehouse now 15+ PB; capacity in a week, not six months

Lowe's

A packaged e-commerce platform

From late 2018

Rebuilt on an internal platform (Carbon)

Deploys in hours; carried 111% e-commerce growth in 2020

Liberty Mutual

New applications, under a serverless-first policy

From 2019

CDK pattern library; serverless first

Build time from a year to three months; ledger at $60 per million transactions

FINRA

Market surveillance analytics

Described 2017

S3-centred lake, storage separated from compute

Years of history queried in minutes or hours

Target

Software in 1,800 stores

Described 2018

Kubernetes per store, central intent via Unimatrix

Same deployment experience for stores as cloud

Chick-fil-A

Software in each restaurant

From 2018

Small Kubernetes clusters, GitOps pull model

A fleet of roughly 2,800 clusters managed from Git

Walmart

Its whole infrastructure

Ongoing

Own cloud plus two public clouds in three regions, plus store edge

One platform layer over all of it

Want a second opinion on this for your system?

Book a free 30-minute call. You'll leave knowing what we'd do first and roughly what it costs, whether or not you hire us.

Six patterns

  1. They started where the old constraint hurt most and the result was visible: Home Depot's warehouse capacity, Lowe's e-commerce, Netflix's single database. Early, visible wins funded the rest.

  2. The core was rebuilt, not moved. Capital One rebuilt 80%; Netflix refused to lift and shift; Amazon chose a new database per workload. Rehosting was used for the long tail, never for the systems that differentiate.

  3. The riskiest systems moved last. Netflix moved billing and customer data at the end, once the platform and team were proven.

  4. A platform made the right way the easy way: Lowe's Carbon, Liberty Mutual's CDK patterns, Target's Unimatrix, Walmart's common layer. Adoption followed convenience, not mandates.

  5. It was a people program. Capital One grew to about 11,000 technologists; Home Depot and Lowe's each announced plans to hire a thousand or more engineers; Amazon retrained its DBAs. None bought a transformation.

  6. Cost was a result, not the reason. Netflix called lower cost per stream a welcome side benefit. The stated reasons were reliability, speed and capacity.

For a company of 50 or 500 people

  • Pick the one system whose limits the business feels most, and modernize it properly first.

  • Retire and rehost the rest quickly so the old environment shrinks while the important work happens.

  • Build a small paved road (two or three templates for your common service shapes) before the second project, not after the tenth.

  • Move the systems that hold money and customer data last, with parallel runs.

  • Budget for people: training, a platform owner, and time for teams to learn the new way.

hoolco Monthly Review

Once a month: what changed in tech that month and what it means for you, from cloud and AI to security and the tools you build with. Every item links to its source. No sales pitch, unsubscribe in one click.

As of October 25, 2026. Provided as is, for general information, and accurate to the best of our knowledge on the date shown. Not legal, compliance, tax, financial or other professional advice. Prices, limits and vendor services change; check current vendor documentation before relying on anything here. Terms of use.