What Happened to ThoughtWorks?

Why has a leading software technical consultancy firm, once renowned for its innovation and positive impact on Agile practices, gradually lost relevance over time?

A repost of my past article on Medium in 2024.
(my most popular article, with 1.5K claps from ~400 readers)

(The share price of 2023–04–12: $2.39. The stock jumped nearly 24 per cent in Nasdaq debut, at $26, on 2021–09–16. What a big difference!)

Last night, my daughter, reading the “Continuous Delivery” book, said, “It seems many things came out of ThoughtWorks”. I replied, “ThoughtWorks had a very good reputation. Some world-class and innovative programmers worked there”. Then I listed, without thinking, a few thoughtworkers (the term they used to describe Thoughtworks Consultants; the years in brackets are their ThoughtWorks time):

• Martin Fowler, who needs no introduction

• Jon Tirsen, (2003–2008), the creator of Nanning, which inspired Dependency Injection

• Bret Pettichord (2004–2005), creator of Watir.

• Jason Huggins, (2000–2007), creator of Selenium v1.

• Aslak Hellesøy ( 2003–2006), creator of Cucumber.

• Simon Stewart, (2004–2007), creator of WebDriver.

• Jez Humble (2004–2014), author of “Continuous Delivery” and other best-selling books on CI/CD.

Then, I googled the ThoughtWorks Share Price, which showed a shocking 86% decline over the past three years!

Its share price is going in just one direction: down, since it floats.

Objectively, ThoughtWorks did quite badly in recent years. This can be reflected in few people mentioning ThoughtWorks nowadays. This made me a bit sad.

Table of Contents:
· My Personal Experience with a great mentor, a ThoughtWorker
· Most ThoughtWorks Products have failed
∘ 1. Mingle, Digital User Stories, lost to JIRA
∘ 2. Failed Test Automation Framework/Tool: Twist and then Gauge
∘ 3. Failed CI Tool: Cruise CD and then GoCD
· Stopping its Ruby on Rails Consulting
· Poor Test Automation and Continous Delivery Consulting around its no-good products

My Personal Experience with a great mentor, a ThoughtWorker

In 2005, I was extremely fortunate to have a chance to pair programming with a ThoughtWorks consultant, a world-class programmer. What I learned in those six weeks transformed my professional career, including

• Refactoring with IDE (IntelliJ IDEA)

• Unit Testing with mocks (using jMock)

• End-to-End Functional Testing (using JWebUnit)

• Using Page Object Model to make E2E Test Scripts easier to read and maintain.

• Simple Design is Hard

• Continuous testing: daily execution of the whole unit and E2E test suites.

• Embrace Ruby, a great language

• User Stories in physical index cards, not digitally

I earnestly adhered to the above advice and valued it greatly. However, ThoughtWorks nowadays seems to have deviated from the above.

Most ThoughtWorks Products have failed

In 2005, my mentor told me that ThoughtWorks only provided technical consulting and guidance. All the libraries, frameworks (e.g. JWebUnit) and tools (CruiseControl CI Server, User Story Spreadsheet) we got from them were 100% free (both in freedom and price).

Of course, ThoughtWorks’ decision to develop commercial products was justified. But, they failed, mostly.

1. Mingle, Digital User Stories, lost to JIRA

“Mingle, an Agile Project Management product, was originally created at ThoughtWorks in 2007. In July 2018, its retirement was announced.” (link)

“We proudly launched Mingle in August 2007, with the world’s first ever digital card wall to support Agile project management. … But now, after 11 years of capturing backlogs and managing iterations, we’ve decided it’s time to retire Mingle.”

But it lost to JIRA.

2. Failed Test Automation Framework/Tool: Twist and then Gauge

Thoughtworks’ two attempts to Test Automation products failed because they went the wrong paths.

Twist (2008–2014)