Cypress vs Playwright – April 2025 npm Trends Download Count Comparison
The numbers confirm the accuracy of my predictions made over four years ago.
I predicted that Cypress would fall when I first tried it out about 7 years ago. When starting blogging on Medium in 2021, I published a few well-received articles on Cypress:
Why Cypress Sucks for Real Test Automation? (2021-01-28)
Why Cypress Sucks for Real Test Automation? (Part 2: Limitations) (2021-02-15)
👎🏼 Cypress vs 👍 Selenium WebDriver Comparison by Example (2022-06-22)
Comments on the claims of “7 ways Cypress is different”. All False, Wrong or Lie (2023-07-08)
Cypress.io is Dying (featured in JavaScript in Plain English publication), 2023-08-02
Of course, I did receive some skeptical or disagreeing comments—along with a few rude, personal attacks.
One comment in particular cited the npmtrends download counts (2022) to argue that Cypress remains strong. I replied: “It is a prediction. Anyone can state the current situation, wise ones make accurate predictions about the near future”. Thanks to him, I knew an objective way to verify my prediction: Cypress will fail.
Having spent nearly 20 years hands-on with E2E test automation and continuous testing, I’ve built a solid foundation for making such predictions. Many of my past forecasts in these fields, e.g. failures of Protactor.js and Cucumber (when used in E2E test automation), have turned out to be accurate. This one about Cypress, after two years, proved correct again.
Trend Analysis
Clearly, the Cypress download trend has remained static, while Playwright’s has grown rapidly—especially over the past year—showing an exponential growth curve.
You may have heard some Cypress testers say, in an attempt to cover their incompetence, “We should avoid E2E testing and focus on component testing instead…” That’s misguided. Playwright is an end-to-end web testing tool, positioned directly as a competitor to Cypress. How could Playwright have such high growth?
“<meta name="description" content="Simplify front-end testing with Cypress' open-source app. Explore our versatile testing frameworks for browser-based applications and components.">”
- Cypress.io website home page’s description.
The demand for E2E testing has always been there; Cypress just fell short and was overhyped.
Let’s compare the download counts.
Cypress download counts:
2023-04-09: 4,674,856
2025-04-13: 5,689,003
Playwright download counts:
2023-04-09: 1,236,519
2025-04-06: 13,544,102
Two years ago, Cypress had 3.8 times the downloads of Playwright. Today, it’s down to just 42%. That’s a dramatic shift by any standard. I predicted that four years ago, when there were no signs.
This also highlights the value of reading expert articles and books with an open mind. A wise testing professional who read my article, 'Exit Strategy for Cypress Testers,' and took action would now be in a comfortable position.
Cypress causes so much damage to E2E test automation (by faking) and the reputations of Test automation engineers, now with fewer jobs and lower salaries (compared to Selenium WebDriver prime time). I hope to see the news that Cypress.io is winding up, the day that real test automation engineers should celebrate.
Related reading: