Selenium HQ Reviews

We are using selenium in our organization for Unit Testing, Smoke Testing, Sanity Testing and Regression Testing.

Reduces Manpower, Testing is become much faster which is helpful for faster Project delivery. 

Selenium supports a range of languages, including Java, Perl, Python, C#, Ruby, Groovy, JavaScript, and more. It has its own script, but it is not limited by that language. It can work with various languages – whatever the developers/testers are comfortable with.

Selenium provides support across multiple browsers, namely, Internet Explorer, Chrome, Firefox, Opera, and Safari. This becomes highly resourceful while executing tests and testing it across various browsers simultaneously.

I have been using this solution for approximately two years.

The stability of the solution has been good, it is reliable we have not had any bugs till now.

I have not connected their technical support. However, there is a lot of community support for Selenium HQ online. If you look on internet you can find plenty of solutions for challenges you are facing.

The initial setup is little complicated the installation could be simplified.

I would recommend this solution to others.

Our company is a customer of Selenium and I work in QA.

One of the most valuable features of this solution is Web Scraping although overall I think every single feature of Selenium is valuable because it's purely API so we can integrate it with any other tool.

In terms of improvement, I think there could be additional readability and abstraction in the solution, similar to the way other frameworks function. The solution could possibly use a more generic framework with more machine learning. If the community or IT companies had the budget, they might be able to work on that but it would definitely improve the solution, particularly in areas such as image comparison and image recognition. Machine learning capability would enable a self-healing capability in terms of locators. Selenium is not a magician, you need to provide instructions in order for it to identify any element in the web. That would really be an advantageous feature. Jason Arbon, CEO at test.ai, demonstrated this in Appium, which derives from Selenium. 

I'd also like to see a one-click install package. For now, I have to use TestNG, then Java, and then build them back. That could be simplified. 

I've been using this solution for almost five years. 

It's a pretty stable solution but we have to be careful and make sure we're using what has been released. We need to know which version to use and make sure it's the latest one so we know it will be stable. 

We are a small setup so have four or five users in the company. I've worked in bigger companies where thousands have used the solution on a daily basis, so it's very scalable. 
It's not the solution that deals with scalability, it's the cloud or a virtual system that enables scaling. If there are 1,000 test cases that I want to commute in an hour, then sometimes the reports can be messy.

We use Selenium forums most of the time and we get most of our answers from there. There are multiple people online providing answers. Selenium does provide a service but we don't use it because we are self-sufficient in that respect.

I have used Coded UI from Microsoft, which I believe will be deprecated some time this year. That was a very good solution. Before that, I also used HP QTP. Both of those are easier solutions than Selenium but also quite expensive. 

The very first setup several years ago was quite complex, nowadays it is simple. Generally we use Maven, so it is simply importing and doing the setup. If I already know the settings, it can be deployed within half an hour. It depends on the system configuration so there can be issues sometimes. Setup can easily be done internally and if there are good community people there could be a one-stop solution layer where installation is one click. Maintenance is only required when there are updates or API changes. 

The solution requires good understanding of Python, Pearl or Java and it's important to use the documentation. It would take a few hours of watching video sessions or reading the documentation to get started. It also helps to have an understanding of how API works, the architecture and how Selenium interacts with browsers. Having the background helps to design something with more features. 

I would rate this solution an eight out of 10.  

We primarily use the solution for automation testing.

Nowadays we are using the solution to allow old solutions to run through it. This way, we don't need to go check functionalities created previously. We just run the automation. In just one click you run the automations through. It saves a lot of time and allows us to focus on other things.

The solution is very flexible; there are different ways of using it. It's open-source and has a lot of support on offer. 

It's excellent for analysis automation testing.

You can use the solution on many different browsers as well.

The solution need to  offer up much more information in regard to performance testing using selenium.

The solution is very scalable. You just have to have an understanding of how you are going to use it before you proceed. We do plan to increase usage in the future.

Technical support is very good; I'd rate it seven or eight out of ten.

We also use ReadyAPI alongside Selenium. We're working to integrate the two to have an overall more powerful performance.

The initial setup was straightforward, however, it does depend on what the organization's requirements are. Deployment can take up to one or two months.

The solution is open-source; you don't have to pay a licensing fee.

We're using the on-premises deployment model.

It is an open-source solution and you are able to build vertically. There is also lots of support pertaining to Selenium such as videos on YouTube. It's very easy to integrate on different platforms.

The solution is constantly being updated and advanced; it's very reactive and in a constant state of change. 

I'd rate it seven out of ten.

The ability to test on all of the major web browsers.

I’ve used Selenium in multiple ways of the years. In one case I was working on a web site that got updated once a week. Mostly by others. Every week we would find out that a page changed that we were not expecting because we had changed included content, not knowing it was appearing on another page. I convinced management to allow me to write a script that went to each page on the production site and compare the page to the stage site. I reported the changes to management so that if they didn’t know about them, they were aware of the change.

On another site, we used Selenium to do basic application level testing. Just about the time I would start to think I was wasting my time, the test would reveal a problem that needed to be addressed.

Can’t think of anything I would change.

I've used it for at least five years.

There's never been an issue with deploying it.

As the various browsers change, tests stop working. But that can hardly be blamed on Selenium. Every problem I've ever had has always been answered by “Get the most recent version of Selenium.”

I've never had a problem scaling the tests.

It is open source, I don’t think this applies. The documentation could use some work but there are books that fill this hole.

I had used TestComplete at one point a very long time ago. What I didn’t like about TC is that it was a complete development environment, which I didn’t really like. Selenium gives me the ability to write my test using the programming environment I am already familiar with.

The initial setup is pretty straightforward, but when you branch into using the Selenium Grid, things do get a bit more complex. It is all pretty easy now that I know how.

I implemented it myself using search engines and trial and error.

It’s free. And relatively easy to get setup.

Learn how to use the “Page Model” approach to creating tests. If you code direct to the Selenium API and don’t using the Page Model, as things change in your web application, it will become more and more difficult to update your test to reflect the changes.

It allows us to verify user flows effectively and very easily. I can use any web browser to assert user flows.

It's helped us get good coverage with functional regression tests, reducing the reduction of manual effort to sign off on every release. We can now do so in just one day to sign off a regression test as compared to 4 day before we had automated tests.

Plus, as the regression runs are invoked on every commit, the tool has also reduced the number of issues introduced by regressions and has highlighted the regression issues on every commit/push. This allows us to get very early fixes and reduces the risks associated with changes and releases.

Overall, there aren't any areas for improvement. However, there are some things you have to do make make effective use of the tool. Make sure the language you select to work with is the same as the language your applications is developed in. Also, make sure you use Selenium Grid to get the test and suites to run in parallel.

We've had no issues with deployment.

Stability has been fine for us.

It scales without issue.

It's open source, so there's no customer service.

It's customer service, so we provide our own technical support.

I like that it is a robust and free open source. There is a lot of community support available, and there are a lot of developers using them. There's good community support.

In terms of improvement, there should be standardized frameworks, which do not exist. Currently, you have to first sit down and look inward. This function is provided by some tools like Katalon.

There should be standardized frameworks to build automation.

I have been using Selenium for five to six years. 

We also use Sauce Labs. We do testing on various versions of browsers, and various operating systems for Android and Apple. So Sauce Labs provided that capability, where we can easily scale up our testing to test on various versions.

The scalability and stability are quite good. It is easily scalable. With a few pieces of code, it's scalable.

We have interacted on the Selenium forums a couple of times but never contacted the technical support.

We also use Sauce Labs.

We set up the entire framework around how the tests would be and different class structures. Then we evaluated how to construct labs and how to title execution and stuff like that. It took us good four to five months of two or three people working on it, to set up a robust framework.

If you are using it for the automation test, automation is generally good. If you have an experienced developer, it will be really good to deploy the open-source frameworks for Selenium.

I would rate it a seven out of ten. 

Not a ten because they don't come loaded with the frameworks or different framework options. You have to build it from scratch. The build-in comes with a lot of networks and libraries and stuff. You still need to build that framework before you start writing the code

They try to come up with new features. I think they're in a very good space.

Selenium web driver - Java.

As traditional load testing we use some expensive tools available in the market, but with the help of Selenium we started automating the browser and done load testing for light application.

Handling frames and windows needs to be improved.

I've been using it for one year.

No issues encountered.

No issues encountered.

No issues encountered.

9/10

Straightforward by using Selenium IDE.

We did an in-house implementation, doing our R&D over the internet.

Hot tool in market. Makes thing easier to use and implement. Demand increases cost.

Since it is an open source. It is free to use. However my company see it as the future of load testing.

You'll have a lot to do when automating many different applications. Just practice.

• It's open source tool (cost effective)
• More Test Coverage through Multiple browser support(Chrome,FF,IE9,10,11,Opera,Safari and Edge(recently)
• Scalable to extend and integrate with other Automation tools like Jmeter & OWASP ZAP, thus enable team to perform Fuctional, NonFunctional test(Performance, Security)
• Can be integrated with CI tools like Jenkins
• Can be integrated with builder tools like Maven & ANT

• Can be extended to cloud environments (sauce labs)
• Multiple platform support like windows, Linux and MAC
• Can be used with different Automation frameworks like TDD,BDD and Hybrid 
• Remote script execution possible with Grid & node setup
• Less Turnaround time with Parallel Executions 
• Easy to Learn & Mastering
• Huge user community base and support
• Support for version control tools like SVN and Git
• Headless Browser support using HTMLUnit & PhantomJS drivers

It is an awesome tool to use with great ROI, we have been using this tool extensively in most of our projects to reduce the manual regression efforts and to prevent production defects. Another beauty of this tool is to extend and integrate with other functional & non functional tools(Jmeter, OWASP ZAP, Appium ..etc.)

Recently we have implemented the Extended Reporting to support rich reporting structure with Charts,Dashboards that would be easy for stakeholders to Analyze and arrive at decisions. 

Compatibility issues with newer versions of browsers (IE & Firefox) and latest server versions, but this can be addressed with some workarounds.

Handling Windows based dialogs (or) performaning Non browser action can bit challenging, so far we are using AutoIT tool to perform this.

Rather QTP or any other Commercial tools, Selenium expects more Techinical(Prograaming) expertise, but other end this feature makes selenium more flexible

We had compatibility issues with some newer versions of browsers.

We've had no issues with stability.

We've had no issues with scalability.