It annoys me because you have to create everything in one development environment and then push it out to the run-time engines, but anything that has the power or capabilities of what OO does, it almost has to be two separate entities. It's easy for people who don't know what they're doing from a logical perspective to really screw things up.

You can build some safety features and everything into it where that doesn't happen, but really primarily what you do is you limit the amount of people who develop the flows. That way you stay consistent.

Simple things are missing – being unable to easily capture flow inputs in an automated fashion is one (I shouldn’t have to write an SQL query for this). Being able to easily debug only certain sections of a flow is another (overriding responses doesn’t cut it). They could also improve on variable tracking within a flow. Implementing a find and replace feature would also be nice, so you don’t have to do it through the XML on the backend. Making a parent flow name available to the subflow it kicks off (outside of central/without going into the database).

The tool's UI needs to be improved. It needs to have better administration features in future releases. 

It takes a lot of resources to implement and support it, which is expensive.

It's hard to tell what might be some areas of improvement because some of the constraints might be access-based or have role-based access, so I don't necessarily have access to them. Obviously, we have administrators who narrow down the privileges of our use.

But what I would like are more functionalities around code release. We have a lot of people who spend a lot of time manually deploying code into our thousands of environments. So to make all that more efficient and easier, there should be more functionalities.

I'd also like to see more enterprise-level improvements with concurrency and scalability. It should be big enough to work with the state that we have, but I think there are some concurrency issues such that we can only do 10 or 20 things at a time.

It is an agent-less tool. It is two-sided. It is good, because it requires less approvals for deploying. But, it gets difficult because wherever you typically want to deploy robotic process automation related use cases will not have an agent sitting on the endpoint. Executing activities and tasks on the box without an agent can become a tricky process. There are some workarounds that can be orchestrated, but it is extra work. 

I would like to see a bit more in the UX. It needs to be a bit more up-to-date. The version we've got is pretty new. It could do with a bit of a web-content makeover.

We use a lot of HPE’s software, the main point being that a lot of the server hardware based tools we used are using very new, integrated web pages whereas the HPE SA and OO are looking quite old, still fat client based and prone to some unfortunate service issues. Maybe we are not using the latest version and these are addressed, it was just a major point I thought was worth addressing.

I'm not quite sure, because the toolsets that are introduced out of the box are very helpful. Once you get that learning curve, when it clicks, it's kind of like a game to me. Once you have the personnel who can really get into it and really enjoy it and make things better, I really cannot find any specific examples for you to say that I wish this was better, because it offers you so many options. You're developing something so it's kind of how you want to do it. There's so many different ways that you can develop your flow, so they have a ton of options for you.

I'm not a developer by trade. I actually come from an operations background. I have a little bit of scripting experience, and when I took this on, we actually had a contractor come in and guide the work. Once he did that, and then once they left, I took over those flows and just thrown out there in the water. Sink or swim. When you have to do it, it was something I picked up in a really quick. Once it clicked, we were off. We just kept going.

You have to have a subject matter expert. You have to have that SME to have that mindset of a developer. You need to have someone with some development experience. Not saying you have to, because it can be learned, but it is easier to have someone to have that experience and to have that mindset. But you do have to have that dedicated resource, I guess you could say to stay on top of it, because if you don't stay into it, once you create these flows, and if they're very customized, some of them in my experience can get quite large. You have to keep up with what's being added, what's being updated and know exactly what's going on, because over time you can kind of forget things that are happening within the automations.

One area is support. The other one is the release of new products that emerge in the market. This tool is set up such that it encompasses third-party integrations that can be slow. I would like to see the introduction of new products to market and getting support for the application as quickly as possible.

• Product installation
• Audits
• Monitoring section

Better documentation and support from HP.

The integration grid between the SA and OL needed improvement, but that was solved. Like in SA studio, we needed to have our workflows and stuff created. Then, using a batch file you had to synch and so on. But that was solved. It is all integrated.

It's not a perfect solution, such as with the customization prompts. They're a little awkward and not very intuitive.

The one feature that would be useful for us would be integration of more automation with other solutions. I think this should be possible with newer versions.

HP recognizes the need to integrate each product into one big suite. They are going to offer a new suite for flow orchestration and other processes, but the current way they bundle the products makes it really difficult for the customer to identify which products they need.

The price is an area that should be addressed because the price is high.

Training.

Skilled Professionals

I'd like to see it in an active-active infrastructure so that if one goes down, the other picks up automatically. Only an active-passive application exists currently.

Also, there are automation tools out there that are more unified -- network, database, servers. Operations Orchestration doesn't do everything and you need to buy other components for other tasks.

In order for our customers to have a deeper integration of Operations Orchestration, it needs to be offered with more cloud options, whether that's public or private or hybrid. I think this would be a great direction for HP to go in.

I'd also like HP to be more proactive in telling us what their roadmap is for this product. It's a market-leading solution so there are a lot of users, but, as with all solutions, greater visibility of what's coming would be a good improvement.

I would like to see more open-source options with this tool.

There are only graphical developments. My technical colleagues prefer to develop as we code, not as we click.

Faster integration with different applications.