OpenText UFT One Reviews

I've found the most valuable feature to be the ease of object identification using the products spy tool.

As with any test automation tool, the ability to run repeatable tests unattended during off hours saves lots of manual testing hours.

I would like to see IDE improvements (collapsible code, being able to open multiple test files simultaneously, having stack trace information). Some of these IDE features have been addressed in the more recent versions.

The solution works for the most part, but the IDE is horrible (although I hear version 12 has a revamped IDE and is much better) and as a result of VBScript being the language, there is no stack trace information available so debugging some errors is not an easy task.

I would like to also see support for other languages than just VBS. Java, Full VB, C#, etc.

I've used it for three years.

The application will occasionally crash or be unable to reach the License Server which causes test suites to fail.

The tech support is pretty good. Compared to Micro Focus’s SilkTest, it is much better. Although I haven’t used SilkTest in about 5 years so it may have improved.

I was not involved in the evaluation of this product. I inherited it.

I don’t really have information on the pricing/licensing as I wasn’t involved in that and wouldn’t be able to comment on the ROI. This solution has been in place for about five years and the tests are pretty reliable so I would think it has a pretty good ROI, but just guessing.

This solution probably wouldn't be my first choice. I have used Silk Test and Selenium. Selenium would probably be my first choice due to the high ROI, reliability, being able to have a IDE choice and support of multiple languages.

• Object repository
• Supported keywords
• API testing

UFT is easy to use so the QA team is not required to have much programming skills. VBScript language is also an advantage that it has.

I think that UFT should support more robust keywords to work with a low number of applications under test.

I've been using it for two years.

It often crashes.

6/10 - I posted questions on the HP forum and mostly received no feedback. I also saw that people post questions and help each other.

I used Mercury QuickTest Pro 8.2 for three years. I still use IBM Rational Robot, TestComplete, and some frameworks based on Selenium WebDriver.

Everything is readable and easy to understand.

We did it in-house.

Open-source automated testing engines are also good.

The Object Spy in UFT is very valuable for spying on controls in our mobile application and viewing their properties and values.

I would like a version that works quicker. Also, a lot of people can't afford it because it's expensive.

I've been using it for two years. Each project is different, and it is sometimes two months or six months at once.

I haven't encountered any issues with deployment.

It's stable, and there's no issues with instability.

We've scaled just fine, and there's no issues here.

Technical support was very helpful and good.

I used different tools and a different solution, e.g. Selenium in a previous company. I didn’t choose this product as it was in place when I joined.

It was already in place when I got here, so I don't know if the initial setup was straightforward or complex.

It is simple to record new automated scripts with the products record function. It makes the process easier since the application converts the recording into code, which you can then alter and configure to your specification.

The process has helped my organization greatly reduce time and cost when it comes to testing a new project. We created automated generic scripts that can test more quickly and efficiently than manually testing.

The application can be buggy at times and takes up a lot of memory on your PC. Occasionally it can crash or not even start, which causes the user to restart their PC. It would be beneficial if it operated more smoothly and didn’t cause PC problems.

I have been using it for a year

Deployment was simple.

The stability of the product was questionable at times, but not enough to where it hindered our work.

It's scalability was very flexible.

Customer service was helpful and knowledgeable

I did not use any different solutions or evaluate any others. This product was determined by upper management.

The initial setup was straightforward. It was easy to load and you're allowed to select the specific plugins you need to use with your applications.

We implemented it in-house since it wasn’t a large-scale implementation. I don’t have much advice or implementation since it's straightforward.

It's great to be able to go from one company to another and each one uses UFT similarly. It's also extremely versatile, in that you can pretty much use it with any type of application.

It allows us to automate hundreds of test cases that would normally have to be manually tested. The tests are also extremely reliable so it saves a lot of time on analysis. That's really the first and foremost benefit for us. We were doing quarterly releases as well as maintenance releases, which is quite a lot. So this solution has saved us a lot of effort and expense.

The knowledge base for getting started isn't terribly deep, so it requires you to have a bit of programming ability to pick it up and use it.

It's deployed without any issues for us.

It's fairly stable, but the problem is that it's not always updated and current. When something new comes up, it takes HP a long time to support it.

Scalability depends on the user. UFT is basically a sandbox and will be as flexible as you make it. So scalability can be high, but there are things that work against it. You're bound by the licensing structure, so in order to get bigger benefits, you have to have multiple copies. If you want to fun multiple simultaneous tests, you have to have the licensing to do that, and that costs a ton of money.

A lot of people are moving away from the big intertools. So people look at products all the time, and every time the budget comes up or every time they tell me to cut expenses or every time they get frustrated with it, a lot of the small-time tools and open-source tools get attention. So I've been evaluating those.

If you're an HP shop, you're probably going to go out and buy it. But I don't think new customers will sign on to replace, for example, Selenium. There are plenty of open-source options, and people who know how to implement UFT already know how to implement open-source codes. So I think that people who are using open source will stick with open source, and people who have significant investment inHP will stick with HP.

UFT is easy to use for functional testing, so for me it’s very important that it can travel across a large range of technologies. We can use the same tool for all the different kinds of automation that we do.

We’re already using LoadRunner and ALM. UFT gets along well with these other solutions, blending together with the ALM suite.

UFT is rather slow in execution, and that’s something that needs to improve. It runs strips rather slowly as other tools handle the same modular strips much faster.

It's not always very stable, but that depends on how you implement it in your organization. We put it on a separate server host in Singapore managed by our guys in Bangalore, so they make sure that they’re always available first.

Scalability is not that important for UFT since it’s not used by so many people at the same time. For us, there's only a few guys performing performance tests so scalability is not a big issue.

Technical support is quite good, though sometimes it depends on who you’re dealing with. Sometimes you get bad luck and get a guy who doesn't know much about it, is new, or is in training, but most of the time it’s all right.

It was rather easy and you really can do it yourself.

Ask a good HP expert how you need to do it and they'll tell you how to do it.

The ability to customize automation using code is the most valuable feature. With the release of 12.5, UFT now includes a LeanFT license which will plug in to Visual Studio and allow usage with C#.

A close second would be the Object Spy, i.e. how UFT identifies objects. Although you can code with CodedUI and Selenium, the object identification tool in UFT is far superior.

Due to cost, my current employer doesn’t use UFT. However, I’ve used it at many different locations in the past. Automation in general will always provide value in test coverage. UFT bridges the coding gap by using VBScript, which is easier for less technical personnel to pick up. This greatly increases the number those people who can use the tool. The competition will require a developer-level skill set to get the same functional benefits.

There are some command-line and other crude methods to integrate UFT into non-HP software suites. This area could be improved, but overall there is little incentive for HP to do so.

I’ve used UFT for just over 12 years. In comparison, I’ve used CodedUI for about six months, and Selenium for only two months. As you may imagine, I have a significant grasp on UFT and what it can do. An argument could be made that with the same experience in the other two competitors, these tools may be on more equal footing.

As with any code base, well-designed and implemented automation code will make for easy maintenance, and a stable code base that will scale very well as the complexity of the suite grows. Nothing will save you from poor coding practices.

I’ve had little direct support from HP. I’ve instead used HP partners to get licensing and support. Specifically Orasi, they have been very helpful in the few support areas I’ve needed. I would rate them 9/10. As an advanced user, there are issues I’ve come across that Orasi wasn’t able to find a solution for. However, they did their due diligence and escalated to HP as appropriate.

It is very straightforward in its install and setup. An extra layer of complexity exists when trying to integrate it with Quality Center, but this is done via some settings and an add-in. The nice thing about UFT is that it’s a mature tool that has massive user-forum support on the internet. Although there is a lot of support for C# (Coded UI) and Java (Selenium), there isn’t as much in regards to the testing side of those languages.

Cost is the biggest issue with UFT. It is not cheap. However, when evaluating the full cost (not just the licensing), I believe UFT is actually a cheaper solution in the end. That being said, seat licenses that are tied to a specific machine (can be moved via support) run approximately $10,000 with tax and associated annual maintenance agreement. A concurrent license runs approximately 17K with tax and maintenance agreement.

From an ROI standpoint, you need to look at the automation effort in comparison to the manual work it would reduce, as well as increased code coverage and a consistent level of testing. In most cases, it will take 2-3 years before the automation suite is significant enough to start seeing its cost even out. Any organization contemplating automation should have that type of commitment to see the automation effort become successful.

I’ve evaluated a few different automation products. Only Selenium and CodedUI come close to the functionality and adaptability that UFT provides. I support UFT as the best solution due to the skill set needed to operate the tool. VBScript is verbose, but otherwise very easy for a non-coder to pick up. Selenium and CodedUI both require OOP languages that are more complex for the tester to pick up.

Overall my main concern, is with resources. There aren’t as many Selenium/CodedUI professionals in the marketplace therefore when it’s time to deal with turnover, it’s very difficult to find experienced automation personnel. That is not the case with UFT. An added plus for UFT is how it handles poorly designed and implemented web applications. I didn’t fully see this until I began to use CodedUI and Selenium.

If this is the first time implementing a solution, I would say make sure to read up on what it will take to implement. Get as much knowledge ahead of time to make it smoother. To hit the ground running, it is best to organize your manual tests so that automation can begin as soon as possible. What test cases are for Smoke tests? What test cases are for Regression? Starting automation without defining the work to be completed will waste precious time -- time you are paying for idle licensing.

The core application that we're using in our company is a bit special and resistant to automation. To properly to automate, it needs a plug-in from the software's vendor. The only technology for automation that is supported is UFT, so we're sort of locked into UFT for our core application.

UFT is the only technology that enabled us to actually automate our core application.

It's fine, but we've had to do workarounds for some things.

We've had no issues with deployment.

For us, we started with version 9 of UFT when it was called QTP, QuickTest Pro. For us it worked, though, like the average software, there are issues, but we always had a good ability to provide work arounds. Regardless, it's been pretty stable.

We've had no issues with scalability.

We haven't needed much technical support as we can sort it out ourselves.

Some parts of it were pretty easy to set up. Most of the issues we had were related to organizational stuff, such as managing licenses, etc.

Make sure you take a good look at your approach, such as whether it's keyword driven automation, etc. Make sure you organize things that you're headed in the right direction because once you implement in a certain way, changing direction may be pretty hard. Determine also how you deal with object repositories, how you deal with sharing information, and how important the reusability of scripts is in your project.