Three Types of Efforts in E2E (UI) Test Automation Clarified

Test Creation (~10%), Test Stabilization & Refinement (30%), and Ongoing Maintenance (~60%). Knowing this will help you understand why many E2E Test Automation attempts failed.

This article is one of the “IT Terminology Clarified” series.

There are three types of efforts or phases in E2E (UI) test automation; I put estimated efforts in percentages in brackets (based on my 17+ years of experience in E2E test automation). Unfortunately, most software teams just focus on the first one. No wonder they failed E2E test automation attempts every single time (by trying different commercial tools or open-source frameworks).

I have already written one article on this topic.

Test Creation Only Accounts for ~10% of Web Test Automation Efforts

In this article, I’ll illustrate those three types of efforts with an example, aiming to make this important concept more readily understandable for readers.

I will use the test case “User Sign up”, a very common one. The test script syntax is raw Selenium WebDriver + RSpec; the principle applies to all though.

Table of Contents:
· Test Creation (~10%)
· Test Stabilization and Refinement (~30%)
∘ 1. The Captcha
∘ 2. Dynamic data to avoid one-use-only
∘ 3. Refactor to use the Page Object Model
· Ongoing Maintenance (~60%)
∘ Timing Related Issues for testing Dynamic Websites

Test Creation (~10%)

The following is the test script (whether created 100% hand-crafted or with some help of a recorder, it does not matter).

it "User sign up with email" do
   driver.get("https://whenwise.agileway.net/sign-up"

   driver.find_element(:id, "email").send_keys("newuser@gmail.com")…