Broadcom Agile Requirements Designer could improve the UI. Other solutions have a much better UI. The new UI should have a new modern framework.

The solution could be more user-friendly. For example, attachments could be icon-based to make it easier for the user to notice them. Overall, it would be nice, from the user-interface point of view, for them to improve it.

In terms of testing or defects, we are able to track it currently, based on each task. But if the testing team could have different access rights, that would really help.

Since we're just beginning to use this application requirements tool, I'm not sure what exactly is missing in it.

The biggest thing for us is the effort required to create the workflows. There's a lot of initial work that goes in and if we could streamline that a bit more, it would be better.

At this moment, I can't think of any feedback which would need action. We've very happy with the product. 

We struggled a bit with the initial setup. It could be easier. 

The solution could improve security and authentication. There are security protocols from SHA level one to SHA level two there's the encryption format as well, 128-bit encryption to 256-bit encryption of the data. This is a new trend that is happening right now. From the communication level and the security level, Broadcom needs to enhance its offering

Adding more connectivity to outside applications and the amount of detail we can export to systems of record could be improved. Otherwise, this is a very solid tool with tons of real valuable functionality.

The new release, CA Agile Requirements Designer 2.4, offers more flexibility when exploiting the APIs of test and defect management tools. CA has addressed this concern with release 2.4.

At present, there is no option for test data parameters from ARD for virtual databases. We have to create them in TDM and push them as well. Virtual database connectivity needs to be improved.

They need to come up with some areas where they can create synthetic data parameters easily from the test cases that have been designed.

The only thing CA does not have today is a testing engine. So they have got the requirements piece, SDLC lifecycle with Rally (or whatever it is called now), but they do not have an engine to house test scripts to really pull together the testing pieces of it like HPE does. Really that is the piece that I would love to see them either acquire, build, or figure out how to incorporate into their test suite, because they have got every other software that you can think of except for that one piece that really is essential for a QA organization.

Aligning models so they look clear and readable without having to move boxes around.

Saving different subsets of test cases, use cases, etc., of a given diagram, for different purposes and providing an easy way to name those subsets.

I was looking into some pre-release testing on the integrations with CA Application Test, and when I exposed the test directly to App Test, I found that the tests which we directly exposed have one set of data that we can directly expose and run in App Test. Once we run it, we come to know whether the test has passed or failed. Currently, when we are running a test in App Test, we are running it with TDM data, so it helps my tests run with multiple sets of data at once. I feel that it is a more efficient process if we are running a test with different sets of TDM data. However, if you take data directly from ARD, it is one test with one set of data, which is getting executed. Therefore, data flexibility is something which I would like to see, along with more integration with App Test.

Also, a template in App Test should be created in advance. This has proven to be time consuming. The process is not fully automated, because there is a lot of manual intervention is required. I would like an enhancement or something from the tool that I can use in a more automated way than a manual process. 

We're actually participating with CA to provide some enhancement requests on new features we'd like to see. I can't think of any specific ones off the top of my head, but I know they are evaluating, and they have already provided us features based on our feedback previously.

In terms of improvements, just licensing. That's a big area.

When you have long flows or models, the display can be problematic, as it necessitates zooming in on the GUI and you lose the full display.

As your model grows in size, it becomes increasingly difficult to get a meaningfull view of the entire model. You might have a view with the whole model, but then the text and flow will be incomprehensible. Likewise, as you zoom in, you lose the entirity of the model. A workaround is zooming in as much as needed and scroll around from there, but one could wish for a better solution.

1. A better alignment to compliance-related requirements and being able to get visibility into those. It is really important for our industry. 
2. Test data: There is Test Data On-Demand and there is the vTDM solution. We would like to see a nice integration between the data and the models. That is really going to help us in terms of CI/CD.

Because of the way it is designed, all the boxes are displayed no matter how many there are. So, the more boxes you create the smaller and smaller they are displayed and you cannot even read it. So it's impossible to design your requirements. If you cannot read the boxes, what's the point?

Then, they need to make it more user-friendly, so that you can navigate from one section to another. Also, being able to automatically create stories. Some of the feedback I received from my team is that you cannot go out of the detail level. For example, if you want to create test cases from this level, it actually stays at a really high level unless you add a hundred steps. And then you have a huge process diagram that you have to maintain after that.

I don't know what improvements to make since we are not yet using this in production. We are still in the preliminary stages of testing. I plan to do more research on it. They're already doing the right thing. They're improving the integration with automation and that’s really important. However, the documentation is not clear regarding best practices, so that could be improved. Since it’s not likely that every automation framework can be easily integrated with CA ARD, it would be nice to have recommendations regarding what changes to make to the automation framework in order to successfully achieve that.

ARD is constantly improving, and CA is always working on adding new features. Integration with Agile management tools can be improved, i.e., mainly test case maintenance and linking test cases to the automation script.

We are waiting for new features that we know that CA is working on. For example, to be able to deliver scripts directly from ARD, and then provide automatic testing. After that, with other tools we have, we could have a complete chain to test our application, beginning with ARD and then to Micro Focus UFT, for example. And without manual design.

There are another two issues, I don't remember them exactly, but CA is working on them, so it's okay.

We are using many tools in our projects, like Tosca, but CA ARD is a very closed tool. It doesn't provide integration with Tosca. The possibility of creating a test case and exporting it into Tosca is not available. Integration with end-to-end automation tools, like Worksoft or Tosca, is not provided by CA ARD as of now.

Although there are many integrations which are already available, like Jenkins for automatic performance testing, where we can actually build batches and all, still, there's room for improvement here which the CA ARD team can work on.

I think it's already coming, but more automation aspects involved with it. There is a tab for Automation, but I think it's not robust. I think that it's going to be a crucial element of the tool.

It could be quicker to use. Because we do so many things with it, sometimes it takes a long time to create a flow.

Most of our requests have been met, but I would like to see connection with RSS or maybe with Confluence.