Ghost Inspector tool review (2022)

Andrii Ievtukhov
5 min readMay 8, 2022
Photo by JESHOOTS.COM on Unsplash

Ghost Inspector — a complete end-to-end testing solution. It makes automated testing for websites and web apps easier — but we haven’t skimped on powerful features. Whether you’re new to QA testing or a seasoned test engineer, we give you a versatile set of tools to achieve your quality assurance goals.

GI requirements

System

  • Windows/Linux/macOS

Tools

  • Chrome / Firefox browser (needed for Ghost Inspector extension installation)

An application under the test

  • A website that is available on a public URL in WEB
  • Local / hidden under VPN website will not be reachable.
  • As a workaround, you can use 3rd party tools (like ngrok or Netlify), but these tools are making your website publicly available!
  • If you want to secure your public Website and test it — you can use HTTP Authentication.

User skills/knowledge

  • CSS selectors — basic level (ids, classes, data attributes, selector specificity, etc.)
  • XPath — basic level (nodes, predicates, paths)
  • HTML — basic level (tags, attributes)
  • Javascript — basic level (variables, conditions, loops)

GI features

1. Record/Edit tests in your browser using a very small and UX friendly extensions (Chrome / Firefox)

2. Maintain tests using a codeless editor

Actually, it is not as “codeless” as you may think. You have to understand all generated steps and in complex scenarios (e.g. working with calendars or random data generation) you will have to write/adjust steps/code using Javascript

3. Variables on test run/test case levels + built-in random data generator

In addition, you can create a variable on the fly and use it through the test. It can be a random email address, name, hash, whatever you want. Everything is covered by the Faker package + it is possible to execute any JS code needed (meaning that you are able to generate literally everything).

4. Mark tests as reusable (importable) modules

You can mark a test as a module and import it into another test. It’s very useful for actions that should be done in many tests, e.g. User Login, Order Placement, etc. Also, it allows you to create very complex scenarios and, which is more important, adjusts a lot of tests very quickly (for example if your login flow has changed — you will have only 1 place where the update should be done).

5. Run tests in different browsers*, screen sizes, languages & geolocations

Only Chrome and Firefox are supported now, but at least 10 latest versions :)

6. Automatic visual testing

Each test has 2 results: test execution result and visual test result. Even if all test steps and assertions pass — the visual test can fail if a lot of differences are found on the screenshot, which is being taken at the end of the test (the % of differences can be configured). It really helps to keep an eye on the UI changes and regressions.

7. Tests scheduling and test results alerting

Tests can be run on a very flexible schedule, e.g. once per workday, each hour, or at a very specific time. This allows you to monitor (and control) the state of your test/stage/prod environment easily and without additional tools.

8. Parallel testing by default

All GI tests are run in 50 parallel threads! No configurations/grids/tests adjustments are needed!)

9. Accessibility Testing

It is possible to add an Accessibility check as a test step. The Accessibility report is very informative and contains detailed information about the violation, its severity, and a link to a rule explanation — everything you should have to create a nice bug report.

10. Extensive external API (TOP feature for flexible automation)

A test run can be triggered by the API call with given configs, which is very useful if integrated into CI flow, e.g. you can run some smoke tests when your website was deployed on the `test` environment and if they pass — the tool informs the team that the website is ready to be tested (or moved further in the CI flow)

GI limitations

1. Shadow DOM might require heavy JavaScript usage

This can be a huge problem in some specific projects.

2. HTTP error statuses can’t be checked directly

It is not possible to detect a 404 or 50x response from the server by status code.

3. Alert & Confirmation Boxes are automatically closed with the `True` option. `Cancel` is not supported for now.

It looks like a limitation, but in reality, in most cases, the `Cancel` flow can be ignored :)

4. Safari/Edge browsers are not supported

Not good for sure, but not critical nowadays for many projects :)

5. Websites under the test should be available on the WEB

No localhosts, No VPNs, No mobiles.

6. No Lighthouse checks so far

It would be great to have these, to be able to monitor website performance :)

Pricing

Free trial: 14 days, 200 test executions, 100 team members. More than enough to learn and try it out.

Small: $99 / month, 10k tests (~333 tests/day), 5 team members.

Medium: $199 / month, 30k tests (~1k tests/day), 15 team members.

Large: $299 / month, 100k tests (~3k tests/day), 40 team members.

Annual subscriptions save 10%.

General feedback

The best alternative to the Selenium IDE I’ve seen so far. Easy to learn, easy to use, fast, reliable, with lots of useful modern features, visual regressions, and accessibility checks.

If you are new to test automation — this is a perfect candidate to start with, as only basic knowledge of CSS/HTML/JS is needed.

If you have a small project — this is a perfect candidate to do E2E tests.

If you have a complex project — this is a perfect candidate for your Smoke E2E tests.

If you want to set up E2E testing very quickly — this is a perfect candidate. 10 minutes — and the first tests are running without any line of code written!

If you want to be more flexible in test automation and execution — this is a perfect tool for you. Run tests when YOU need, run ONLY tests YOU need, change any test data without repositories/test-runners/configuration/long CI checks, and forget about browser/web drivers/tools updates.

If you are QA Lead / Senior QA Engineer — explore GI API. It allows you to create very cool and flexible test executors and collect all needed reports and metrics.

The only thing that I’m missing is a Lighthouse integration — would be great to have it in the future to be able to detect some issues in SEO / site performance.

Anyway — it is 5/5 and a tool for every QA Engineer working with Websites in 2022!

--

--

Andrii Ievtukhov

QA Lead. Love Ukraine, my Family, Javascript and Quality Products.