How Nationwide Scaled DevOps: Lessons from a Fortune 500 Insurance Giant

Company Overview

Nationwide is a U.S. insurance company that ranked 68th on the 2017 Fortune 500 list with revenue of $40,074.1 million.

DevOps Case Study

Nationwide is featured as a case study in the *DevOps Handbook* and is a frequent speaker at DevOpsDays and vendor events. The article analyzes their November 2017 San Francisco DevOps Enterprise Summit talk titled “DevOps Handbook Experiments in Accelerating Delivery – Nationwide.” The talk video is available on Tencent Video.

Key Insights

The author notes that the live presentation reveals a more compelling story than the book translation, showing how a large financial‑industry organization achieved impressive DevOps results.

DevOps Climbing Metaphor

The team uses a mountain‑climbing analogy to illustrate DevOps adoption, dividing the journey into three stages—Base Camp, North‑Slope Camp, and Summit—each populated with practices drawn from the *DevOps Handbook*.

IT Organization Scale

Nationwide’s IT organization exceeds 5,100 staff, with over 2,600 developers and testers. More than 200 product‑development teams serve 23 business units, supporting a complex, multi‑layered structure.

Organizational Structure

The structure includes an I&O (infrastructure & operations) team, shared‑service teams, a governance group of CIO‑level stakeholders, and business‑unit leaders. This complexity creates challenges for cross‑BU continuous‑delivery pipelines.

Strategic Foundations

Agile – adopted 10 years ago to deliver high‑quality software.

DevOps – introduced in the current year to improve speed, efficiency, and risk reduction.

Lean IT – aims to scale Agile and DevOps practices enterprise‑wide.

CMMI – uses industry maturity standards to assess and improve software development capability.

Timeline of Experiments

2015 – Simplification: early “DevOps Cabin” using IBM UrbanCode for deployment.

2016 – Organization: leadership support enabled improvements in development model, lean IT, and supply‑chain management.

2017 – Experiments: focus on Minimum‑Marketable‑Product (MMP), programmer experience, and systematic validation of Handbook practices.

2018 – Acceleration: scaling tools to 200+ teams, adopting Google SRE concepts, treating delivery pipelines as products.

Team Interaction Model

All work follows PDCA and three core principles: speed above all, practitioner‑driven success, and end‑to‑end value‑stream scope. Double‑mode experiment teams submit ideas to a shared DevOps backlog, with leadership, platform, and governance teams ensuring alignment and scalability.

EPICs and Experiments

Work items (EPICs) are directly traceable to Handbook chapters. Examples include:

Automated testing: reducing a massive test bed, pairing developers with testers, and enforcing a zero‑tolerance policy for failed tests.

Performance testing: moving from a 90‑day turnaround to a 2‑hour cycle by handing scripts and execution rights to development teams.

Developer Experience Enhancements

The team streamlined tooling, introduced Rocket.Chat for communication, and deployed a chat‑bot that triggers deployments with a simple “deploy” command, dramatically cutting manual steps.

DevOps Climbing Guide

A one‑page handbook visualizes the climbing metaphor, providing new practitioners with clear guidance.

Lead‑Time as the North Star

Lead‑time is the sole metric used to gauge DevOps impact. It is measurable across 200+ agile teams, understandable at organization, department, and team levels, and drives continuous improvement with real‑time daily tracking.

Support Model

Support teams provide coaching, skill‑building, and platform services to ensure 200+ teams can adopt DevOps at scale. Practices include immersion pairing, monthly DevOps dojo sessions, and treating pipelines as products.

Success Story: Hurricane Harvey

When Hurricane Harvey caused a surge in vehicle‑insurance claims, the IT team collaborated with the business to cut 40 % of claim‑process steps and delivered the solution within the same business day—reducing a typical 90‑day change cycle to hours without added risk.

Conclusion

The case study provides a valuable reference for large‑scale DevOps adoption in the financial sector, highlighting organizational structure, metric‑driven improvement, and concrete experiments that other enterprises can learn from.