I work in an automation Center of Excellence in health care, and I utilize it for everything and anything that is appropriate in health care. We have approximately 41,000 hours worth of automation that operates and handles tasks ranging from assisting with HR operations, such as leave of absence, to medication monitoring and much more. Our portfolio is quite expanded.

We are also jumping into the Co-Pilot world because we want to start embedding it into Teams, SailPoint, and some of the other tools. 

We are also looking at a unique use case. We have discharges, and there are certain types of prescriptions that are just extremely specialized as part of discharge. We have medication contracts. One of the things that we have going on in our organization is a buddy system and then an official request. The buddy system is essentially prescription denial appeals. We would like to use AI to search our contracts to figure out what is the standard for these medications and what are the denials and appeals that we can have prewritten and ready to go.

It was a response to COVID that prompted us to dive into scalable tools. We needed scalability and the ability to perform 10,000 or 1,000 or 200 tasks at any moment. No other software that we have had could provide such scalability and is built in a modular way.

It is unique and almost invisible. You do not realize its impact until you discuss it with users who provide positive feedback. We build automation, support it, and deploy it. When we meet with our users later, they have such good feedback about how impactful these things are to their path. For a nurse, not having to repeat a task 10,000 times has made a significant difference. Especially during COVID, it was truly helpful on the front line. Now, I also employ it with other powerful medication tools.

I was personally able to see its benefits as soon as I jumped into software. It was one of the most impactful and amazing tools for my day-to-day life because, at any point in time, I could design something, modularly build it, and deploy it. It can handle such amazing tasks, including things like templates and design and all other materials that we have for communications or activities, and produce them for me automatically. For me, personally, it made a huge impact right from the get-go, but as an organization, we definitely had to learn how to make choices on our use cases.

It has absolutely saved us time and costs. The whole point of the software is to develop something that at least on a use case basis provides value. We have many years of learning how to do our program. Over time, we have also been using it for other things, such as compliance. We have definitely evolved from just saving time or money to using this solution for everything we can use it for.

I have just built my first Co-Pilot, and it is significant in how that part functions. I am also exploring the API task tools, which appear to be quite significant, but for the most part, the forms and the capability to integrate data from other software represent the specialty.

One thing I learned with Co-Pilot is that it is very static. For instance, when dealing with tables and their headers, I must program them in a form. If my build changes, I have to return to my form and update the table titles. Features like looping are missing from Co-Pilot. If I want to perform a task five times in a row, I have to rerun Co-Pilot five times just to achieve that. This is an aspect I considered as I explored possible changes and improvements. Other than that, it is among the best software tools and organizations I have worked with. I find it challenging to suggest much beyond continuing the good work and expanding on it.

I first heard about it back in 2021. This will be my fourth year using it.

I had an issue lasting for the better part of three years, and it was quite severe. I never lodged a ticket because whenever we attempted to resolve it internally, the problem temporarily ceased. Each time I ran my Digital Colleagues, I encountered a strange error on my screen, which persisted for a couple of years. However, an update corrected it, and it is no longer an issue. 

The only current challenge involves a discrepancy in table columns. In my support session, we could not figure out why my table was off by one column. I had to have a blank column zero for my table to align correctly. These minor issues aside, the tool now functions more stably. 

Previously, inconsistent search behavior occurred in the ServiceNow search bar. It was puzzling because it worked differently every single time that we searched in that search bar, but that issue has been resolved.

Scalability is the most remarkable feature of this tool for me. For example, we have a recognition system for automation. I designed a template for various achievements by citizen developers, such as completing a class or build. At first, it was not a big deal to just do it yourself because you are only doing a handful of templates, but the last round had 34. Initially, creating a few templates was manageable. With time, however, it escalated from 5 templates to 34. We are now using this tool multiple times weekly for template generation. Eventually, I aim to release it to the entire automation team so that leaders can readily recognize achievements based on team activities. 

Scalability is impressive; whether handling five or a hundred million templates, it remains easy to design and build. The tool's intuitive nature contrasts with other tools, which are less intuitive for building and scaling, explaining my infrequent use of them.

I have contacted their support for technical issues with the software as well as personal build support.

The response is very quick. Within a couple of days, they are communicating and establishing connections to set up solutions. In my case, I meet with my customer relations every Friday. We get prompt responses to any issues I raise. The support is fantastic, even when dealing with complex situations due to the complexity of some of my builds. While not all problems are solved immediately, this is typical for complex tasks.

I very recently had one of the best support help sessions that anyone could have ever asked for. We were doing something really complex in Co-Pilot, and it was something that was completely unique to even our internal development team. They helped me solve that, and they did it in such a great way. It was a ten out of ten for connecting to someone, taking the time, and doing all of the work. However, we have also had some support tickets in the past that have been just okay. It has been a mixed bag on that front. I would rate them an eight out of ten because they do try hard to connect and support us even though we are doing some complex work.

Positive

This predates my time, so I am not fully sure. I do know Microsoft automations were used before transitioning to Automation Anywhere. Selecting a vendor was a part of the process when we started the program, and Automation Anywhere was chosen.

The decision for the software had already been made. I was only involved in meetings to discuss how we were going to deploy programs, design things, and create standards. 

I believe it took six months from deciding to go live with the software to deploying automations. It did not exceed six months to get everything up and running, though there might have been a year of planning. Once I was set for launch with everything in place, the transition from purchased to used tenants happened swiftly, possibly in less than six months.

Initially, I used strategic resourcing for implementation. At the time, I had an engineer, a technical lead, a technical developer, and an architect—four individuals in total—who implemented all my environments. Four environments were set up, following designs, standards, and rules. This setup involved comprehensive program design, focusing on transitioning and using different control rooms and establishing standards.

We support the entire tool, and we support the automations which require lots of maintenance and support. That is only due to user design or systems changes or some of the elements that are out there. We have a whole support program that goes along with our automations. As for the tool itself, it is just like any other software tool. It is certainly more stable than something like Microsoft where you constantly have to go back and correct issues or incidents. Our incidents are with bots, not with the software.

My job involves citizen development. I work in the field of helping people connect to this technology and use it. In terms of ease of use, the software is not hard to use, but things are presented to users in a new way than what they are used to. I had to make people start with basic programs and learn how to do a simple process. It is a little bit of an advanced tool. It is easy to understand for me and I love it, but it is definitely a step above entry. Before I instituted an entry course, the learning curve was a little bit steeper. We would have individuals coming to our four-day course and then come through and do lots of support. When we taught them the basics of how the automation software works and then brought them into this advanced tool, it was very quick for them. Our results were much better, and we have those individuals building regularly.

In terms of integration, what was interesting to learn was that everything requires a configuration. For an organization like ours, it involved figuring out how we should branch, how we should grow, and how we should engage other teams to get those configurations implemented so that packages will work as intended. That was an interesting learning curve for us. We had to adjust how we train, talk about packages, and use the software to make sure that we had the right integrations and setups. A great example is ServiceNow. They have the packages for ServiceNow built in. If you have all the access to the information and all the things you need, you are in good shape to have some really awesome automation. If one of those pieces is not in place, you have to think a little bit about how you are going to access that.

Automation Anywhere is a ten out of ten. The company is impressive, especially with its new initiatives like Agentic AI. In five years, it is going to be amazing to witness their advancements, particularly as AI continues to play a significant role.

I have used it for multiple use cases across various domains. I started with automating use cases in the utility industry, such as customer billing and processing meter readings. Those were the initial use cases for which I used it. Currently, I am using it for a transport company in the U.S., where we process their invoice documents, perform three-way matching, manage their supply chain records, and handle daily billing bots. All these implementations have been successful.

For the transport company, we have identified a use case. Their ticketing team is manually processing all the incidents that occur during that day. We are now collecting all the documents and information about previously resolved tickets. We are creating a knowledge base out of it, and we will use that along with the AI agents that Automation Anywhere has. We will create an AI agent that will help take care of tickets by learning from the knowledge base. We are working on the PoC, and within the next quarter, we will be able to implement that.

We have implemented a Generative AI solution using Automation Anywhere and ChatGPT for email management. We implemented it for semantic analysis and sentiment analysis. One of our clients wanted us to create a bot that automatically fetches all the reviews from all of the social media platforms, including Twitter, and create a knowledge base within their own platform or database, whether the review is positive, negative, or neutral. For that, we leveraged the ChatGPT plugin along with Automation Anywhere. We created a bot that reads all their reviews, does semantic and sentiment analysis, and accurately describes them as positive, negative, or neutral.

We have not yet used Automation Co-Pilot, but we plan to implement several use cases. One use case involves integrating AI agents with Automation Co-Pilot for conversational AI. Users can input queries, and AI agents will run prompts to fetch required details directly in Automation Co-Pilot. This is particularly beneficial when the process SMEs need to verify post-bot run actions and address any data issues. Currently, the RPA support team is asked to rerun bots. We aim to enable SMEs to prompt bots for reruns, reducing response time from different time zones from five to six hours to immediate actions, significantly boosting productivity.

By implementing Automation Anywhere, we wanted to reduce the manual effort required for various processes for the clients. Previously, the client's accounting team had to collect all the bills, and based on the vendor, upload those to a particular portal. They daily receive bills for more than 250 vendors. We have now automated the process. There is no manual intervention. Earlier, this effort used to take more than 300 hours per month with more than ten people, whereas now, it does not take more than 50 hours per month, and they have only five people.

We have done integrations with Oracle NetSuite and Google APIs. We have done automation with ChatGPT and Claude AI. We are now leveraging Document Automation along with AI agents. We have also done integration with Anthropic. The integration is quite smooth and seamless. 

Document Automation is a game changer. Earlier, with IQ Bot, we achieved partial automation—around 80% of the process was automated. With GenAI and Google Vision OCR now integrated into Document Automation, end-to-end automation is achievable, even with unstructured documents.

The AI agent automation is noteworthy. Everyone is moving from RPA to agentic process automation. I had a chance to delve deeper into it at the Imagine event in Bangalore in 2024, where I engaged hands-on with AI agents. We are exploring use cases to create our own AI agents, which is exciting.

The generative recorder and BPMN converter features are also impressive. If we have recorded the complete process flowchart in the .BPMN file, uploading this file into Automation Anywhere results in a complete bot framework, reducing development efforts by 30%.

Additionally, the generative recorder's ability to handle changes in web portal elements ensures process continuity and notifies developers of updates, aiding in environmental change understanding. If any element changes on the existing web portal on which the bot is currently working, the generative recorder can record and capture the action. The bot will automatically understand at the run time that the particular object has changed, and it will automatically take actions to complete the process and ensure that it does not fail because of that one change.

AI agents need further improvements. AI agent technology in agentic process automation is still evolving. We have only scratched the surface, and there are many potential use cases yet to be explored. As technology advances, I believe it will enable us to implement more comprehensive automation solutions.

Document Automation has issues related to speed. When I compare Document Automation with IQ Bot, IQ Bot is faster than Document Automation. My understanding is that they are still evolving Document Automation, and this issue of slowness will be resolved in the upcoming .35 version, so we are good with that.

They are working on signature identification and validations for handwritten documents. Currently, we are able to extract data out of scanned but digital images, but there is still scope for improvement for handwritten documents.

I started using Automation Anywhere in 2015, so it has been over eight years.

It is stable. I would rate it a nine out of ten for stability.

I would rate Automation Anywhere a ten out of ten for scalability.

Our clients are small to medium organizations. We have seven people working with Automation Anywhere.

I would rate their support a nine out of ten.

Positive

We have completed several migration projects, moving existing bots from platforms like Blue Prism, UiPath, and Power Automate to Automation Anywhere. This was due to limitations with these platforms, such as Power Automate's compatibility issues with legacy systems, which resulted in partial automation.

Customers seeking end-to-end automation have found Automation Anywhere to be the best tool. We analyzed all bots and processes to determine that Automation Anywhere met all requirements. We converted their process documents and redeveloped bots in Automation Anywhere, resulting in successful end-to-end automation. 

Licensing costs for Blue Prism and UiPath are higher compared to Automation Anywhere, making the latter's ROI more appealing. 

Automation Anywhere is currently leading with GenAI and AI solutions. UiPath is also good, but they are still defining its market position. They are a bit confused about whether they want to project themselves as market leaders in AI, RPA, or agentic process automation. Automation Anywhere has already released five to six features directly connected with AI, not just for the customers but for the developers. Blue Prism recently came out with the cloud version. They are still exploring how they can leverage more AI into their platform. Power Automate works great with Microsoft products, but it fails badly outside the Microsoft ecosystem.

Automation Anywhere has enhanced a lot since we first started using it. At the time, it had the client/server architecture. It was not on the cloud. We had to manually install Automation Anywhere on each and every desktop. In 2019, they shifted to the cloud, so we just have to install the bot agent on any computer we want to use as a device. It does not matter whether it is a bot creator or a bot runner. That has improved the installation setup and reduced the configuration time by 80%. Previously, we had to manually install the client and control room and configure them. That was a tedious task, whereas now, because of the cloud, its deployment is very simple and user-friendly, and it has reduced the earlier effort by 80%.

Almost all clients use cloud deployment, but a Korean client has an on-premises setup because they are very diligent about the data. Cloud deployment is easier than on-premise deployment.

Cloud deployments typically take a few hours, around four to five hours. On-premises deployments can take up to two or three business days due to customer environment restrictions and infrastructure limitations. If conditions are optimal, the setup can be completed in one day. For the cloud, deployment time is much shorter, taking only about four to five hours.

The updates happen automatically for cloud deployments, whereas for on-premises setups, we have to download the new release files from the portal and manually do the updates.

In terms of bot maintenance, currently, our team is working for three customers, and there are 105 bots. Our team of seven people is handling these 105 bots. We are doing 24/7 service and daily spend four to five hours on average for bot monitoring and support.

We have one person for the cloud deployment. For on-premises, we have two to three people.

It is not cheap, with costs ranging between 700 to 800 dollars per month. While Power Automate is cheaper at 15 dollars per month, Automation Anywhere offers superior features, user-friendliness, and smoothness. Many customers are transitioning from UiPath and Blue Prism to Automation Anywhere, acknowledging its value. 

The pricing is justified and on-point. There are bulk packages that include licenses, Automation Co-Pilot, and Document Automation. Both Automation Anywhere and partners like us offer discounts.

For non-technical users, I would recommend the citizen developer program on Automation Anywhere University. They have a training portal where people can register themselves and do the learning paths. There is a learning path for new developers or people who do not have much technical knowledge but want to start their automation journey. They can go ahead and do that citizen development training and complete that learning path to understand it in a better way.

It takes about six months to learn it and start using it. The best part about the training is that you can also do hands-on training along with it. You can simultaneously learn and implement it. Six months is a good enough time to get your basics cleared and start your automation journey.

I would highly recommend Automation Anywhere, but the recommendation depends on the customer's specific requirements. For example, for automating tasks strictly within the Microsoft ecosystem, Power Automate will suit. However, for incorporating third-party applications like Oracle NetSuite, I would recommend Automation Anywhere due to its broader capabilities beyond Microsoft applications. Automation Anywhere easily accommodates diverse market use cases to meet customer requirements and scale effectively.

Overall, I would rate Automation Anywhere a nine out of ten.

I am working as a solution architect at Boston Scientific, and I help other teams automate their manual processes using Automation Anywhere as the primary RPA tool. 

The business cases come from multiple teams, such as the HR team, service management team, or supply chain management team. We identify the manual processes that are the best fit for automation opportunities. 

Initially, we review the entire process to understand how the teams are completing their tasks manually. Then, we assess how much we can automate using Automation Anywhere, the features it provides, and how suitable it is for the process. We then convert that particular manual process into automation opportunities.

We are on the path of implementing GenAI for digital transformation. We are doing demos of the AI agents to implement GenAI across all the verticals at Boston Scientific. We are figuring out the right opportunities and use cases.

With chatbot integrations and AI agents, we are able to create responses by extracting the data from the database, massaging those replies, and replying back to the customers. 

We just need to know the end goal to automate a particular process. In Automation Anywhere, we have multiple commands to automate from Excel or record the UI. A person needs to just drag and drop the commands and fill in the data.

Automation Anywhere supports various types of integrations. We have integrated SAP, Excel, Python script, and JavaScript. We can easily integrate them with Automation Anywhere.

We have many opportunities from multiple teams. There are requirements to automate Excel, the UI, or their scripts. We have various integrations with technologies like Python, Java, or macros. Automation Anywhere is perfect for integrating those technologies.

Automation Anywhere has undergone drastic changes over the past five years, transitioning from version 10 to A360, including desktop-based and cloud-based options. The tool is now more stable, with more opportunities and new commands, like recording for the UI or advanced Excel features. 

For beginners learning Automation Anywhere, there should be predefined templates to start creating a task bot. Instead of dragging and dropping action commands and doing error handling, there should be predefined templates that they can directly use. It will help maintain code uniformity across the organization. Besides templates, most features are already available in Automation Anywhere to help automate a manual process into an RPA opportunity.

I have been using Automation Anywhere since 2016, so it has been nine years.

It is stable. I would rate it a nine out of ten for stability.

I would rate it a nine out of ten for scalability.

We have multiple teams in India, the USA, and Costa Rica. We are working from multiple locations. We are a team of 35 users.

The technical support is excellent. I would rate it a ten out of ten. We just have to raise a ticket, after which we receive a call, email, or ping from the Automation Anywhere team. We then provide our queries or show where we are getting stuck, and they provide us with the solution.

Positive

Previously, I worked with UiPath and Power Automate. However, I find Automation Anywhere to offer a better experience.

The setup is easy because all updates and upgrades are synchronized with the newer versions.

We have a dedicated support team that helps deploy the bots. We raise the change requests, which are approved by a technical person and the release manager. After all the approvals and validations are in place, the automation bot gets deployed to the new environment.

The time required for the maintenance of bots depends on the requirement or process. We can see the status of all bots in the control room. It is simple.

We have a completely dedicated team to track bots. We have more than 100 bots running in the control room.

In terms of ROI, we have a good amount, which amounts to millions of dollars per year.

I am not responsible for the pricing part, but I know the pricing is competitive compared to other tools.

It is relatively simple. It is a platform where you get all the commands for integration and other functions to automate your end-to-end process. It is easy to learn and cloud-based. The control room is easy to understand. All these features will be helpful for automating or starting your RPA journey. 

I would rate Automation Anywhere a nine out of ten.

We automate many tasks in our college laboratory using Automation Anywhere. We offer a course on Automation Anywhere. Students add it and complete many jobs using our lab systems. Automation is the future of technology. We want our students to have its knowledge. As a professor, I teach my students how to use Automation Anywhere, and we build many things for them.

Automation Anywhere saves cost and time. It helps to minimize human labor.

Automation Anywhere is very easy to learn. I learned it by watching video tutorials and documentation. It was easy for me. It took me a week.

For integration, everything is available inside the Automation Anywhere software. We can integrate it by using extensions or plugins. We have integrated it with Microsoft 365.

Automation Anywhere has made us more productive. No data got leaked from our software, and it is safe and secure.

Screen scraping is easy. I can easily collect data from any website using Automation Anywhere, creating and saving that data in an Excel or CSV file.

For the tasks that I do repeatedly, I can record processes using Automation Anywhere. I record mouse and keyboard actions and create a workflow for those activities. This minimizes my workload. The recording option is very useful. 

It has AI-powered capabilities that are very beneficial to me, along with significant analytical and reporting capabilities. It provides real-time data.

Logging into Automation Anywhere takes some time. I use it frequently, and each login currently takes some time. A reduction in this time would be helpful. When we enter the automation tool, it sometimes takes a while to refresh the page or perform drag-and-drop actions. Other than this, I am satisfied with Automation Anywhere.

I have been using it for more than one year.

It is one of the most stable and reliable RPA platforms. However, the stability also depends on deployment, system configuration, and use cases. If the UA of any application changes, the bot may fail unless updated.

I would rate it a ten out of ten for scalability. It can support thousands of bots at a time.

We have 5 departments in our college using Automation Anywhere in the laboratory. A single class contains about 138 students, and in a department, there are up to 10 classes.

Their support is very useful. If any doubts arise, they clear them and provide training. We get good service from Automation Anywhere. 

We also have a good knowledge base and a large community forum. If I have any doubts, I can ask and get answers from experts. We also have dedicated account managers and technical experts to solve our problems related to Automation Anywhere. 

Positive

I have used UiPath. Automation Anywhere is a better and easier option. It is powered with AI features, producing intelligent bots. We can generate more processes using Automation Anywhere. UiPath is more user-friendly, but Automation Anywhere has IQ Bot which is very useful for us.

Its deployment is easy. Usually, we sign up for the cloud and choose a subscription plan. We then set up our control room and configure our bots. We develop and deploy them directly from the cloud. It is easy for us.

Automation Anywhere does not require any maintenance from our side, but the bots require a little bit of maintenance. We spend almost an hour on bot maintenance. We see how it is working. If anything goes wrong, we get informed and we then check it out. If there are any errors, we might have to spend more time, but in general, one hour per week or 15-20 minutes daily is enough.

In our lab, we buy it at a monthly price and offer it to every student. The students using it give some amount to us. As a return on investment, we have achieved a 250% ROI in six to nine months, and sometimes it increases up to 380% percent. Our lab monitor collects this data and gives it to us. We calculate and give the final result to the principal of our college. 

It saves us a lot of time.

It is affordable for us.

Automation Anywhere is good. It is perfect to use. The students like it a lot, and it helps with so many purposes.

I would recommend Automation Anywhere to others. It has good scalability and stability. Bots are resilient and reliable. We can easily integrate it with the cloud. Automation is the future, and Automation Anywhere is the best.

Overall, I would rate Automation Anywhere a ten out of ten.

I have been using Automation Anywhere to automate repetitive tasks across various domains and organizations. We have developed a lot of automations in domains like finance, HR, and supply chains on a daily basis. Basically, we create bots using Automation Anywhere that replicate human actions. 

Currently, we are working on a business case involving agents who respond to phone calls, providing order dispatch status. The average call time is five to six minutes. By using AI, we are creating a chatbot so that a customer service agent can input details and quickly provide all the customer-related information. This reduces call times from five minutes to one or two minutes, which is something we are currently working on. We are also exploring AI use cases for things that are not structured or rule-based.

Previously, we were using legacy automation methods. We were using different platforms to automate the system by writing scripts. You get the best value from automation when you automate a process end to end. In a business process, multiple applications might be involved in completing the processes. Automation Anywhere provides better integration with applications. We can integrate multiple applications at the frontend or backend and complete our processes. Previously, we could automate only a few parts because of limitations in supporting various technologies, whereas Automation Anywhere helps to integrate all the applications and automate a process end to end.

I have integrated it with SAP, Salesforce, ServiceNow, Jira, Tableau, and a lot of Microsoft platforms. We have also had integrations with database and mainframe applications. The integration is seamless. They have packages for particular applications. If you want to work with Salesforce, they already have a package for Salesforce. You just have to connect Salesforce with a certain set of details and provide the API details. It is seamless and within a second, you get connected, and data starts to be exchanged with the application.

It saves headcounts, improves customer experiences, and provides time and cost savings. In some projects, millions of dollars are saved annually from a single bot. The response time for certain customer scenarios has also been made efficient.

The way it integrates with subsystems is impressive. When automating business processes, there might be multiple systems in use. Automation Anywhere provides various ways to connect, such as by directly using backend APIs or using the frontend. This is incredibly useful.

They are constantly enhancing its capabilities. They have already included support for GenAI, enabling easy LLM training. You can train LLMs without putting in a lot of effort. You can train a model very easily. If you do the same thing manually from scratch, it will take a lot of time and you also cannot be sure of the accuracy. They are enhancing their capabilities in the field of AI to provide efficiency and accuracy.

It is a low-code/no-code platform. If you have a little bit of technical knowledge, you can easily design a bot. It is a basic part of software development. It is easier if you are aware of a few things such as writing up a logic, loop conditions, etc. Someone having the basic knowledge can easily adapt to this tool. It is quite simple and straightforward with drag-and-drop actions. It also provides information about a particular package or action you are going to use in your automation. If you are not familiar with the technology, you can explore Automation Anywhere University where they provide training. For anyone eager to start with it, it is quite easy.

Automation tools have a lot of dependency on the environment, particularly on machines. Reducing this dependency could increase success. The product has entered the GenAI space, so it is easy to implement. It will keep improving over time. However, making the product more lightweight by reducing its dependency on infrastructure could greatly help in the long run.

I have been using Automation Anywhere for the last six years.

The stability rating is a nine out of ten. They are improving stability daily. If you had asked me four or five years ago, I would have rated it four or five, but they have significantly improved. Currently, nine is accurate, but there is always room for improvement.

It is being used globally. We are a global team with a presence in six to seven different countries. We have about 10,000 employees. We have business processes in different countries, and everyone is using it. We have implemented 140 to 150 automations, and a lot of people are involved in developing them.

I would rate it an eight out of ten for scalability.

I would rate them a nine out of ten. They have been kind and supportive. Whenever we need help, we can reach out to them, and they help us out.

Positive

I have worked with major partners like UiPath and Blue Prism prior to Automation Anywhere. I also worked with UiPath last year.

When comparing them from a developer's perspective, I prefer Automation Anywhere. Its canvas is user-friendly with options to switch between code view and list view, which other tools lack. As a developer comfortable with list view coding, transitioning to UiPath without that feature would require adaptation, whereas transitioning from UiPath to Automation Anywhere is smooth due to the flow view or list view option. Automation Anywhere already provides flow view development, simplifying adaptation. I appreciate this flexibility. I am also working on Power Automate, which is generally flow-based. Automation Anywhere offers convenient switching.

In UiPath, if you want to work with Salesforce, the package will not be there by default. You have to import that package, which is not the case with Automation Anywhere. Importing packages, like for Salesforce, is straightforward, and often available to team members. The best feature is having access to packages imported by team members.

It is straightforward. 

When it comes to bot deployment, a bot takes two minutes from one environment to another. Moving code between environments also takes only two minutes. However, there are sanity checks. These checks apply to all tools used in automation. Deploying code takes two minutes. We have automated this process, allowing code transfer from test to production with a single approval click. It is seamless.

Bot maintenance is not a tedious task. Once an update is done, we push that into a test or development to check whether it is impacting our automation. In more than 90% of scenarios, it does not impact. In a few cases, there is some impact with the update.

Before automating a use case, we assess its impact in terms of efficiency, error reduction, and cost savings.

Automation Anywhere has helped achieve around 50% savings in some scenarios. The error rate has been reduced by 80% to 90% in processes. Errors have been successfully eliminated from the process.

In terms of cost, we participated in projects achieving two to three million dollars in annual savings.

If you are a non-technical person, you need to become familiar with basic programming concepts such as conditions, looping, etc. Automation Anywhere is easy to adapt to. The UI is simple. If someone is coming from a different technology and wants to develop something in Automation Anywhere, one or two weeks are enough to get started with it.

We are getting trained on AI, document automation, and the Copilot system at the organization level. We are also exploring a few other business cases for implementing GenAI. We have some unstructured handwritten documents, and we are trying to assess how efficient it would be to automate it and how easily we can create a model for a certain set of documents for extracting the details.

I would surely recommend Automation Anywhere to other users. For business professionals, it automates repetitive tasks and enhances efficiency and cost savings. For technical individuals, it is easy to use and adapt. 

I would rate Automation Anywhere a nine out of ten.

We have multiple entities and businesses in our organization, which leads to various use cases. Specifically, there are numerous SAP use cases where we perform automation. Additionally, there are use cases with application automation and document automation.

By implementing Automation Anywhere, we wanted to automate repetitive tasks done on a daily or weekly basis and utilize the time saved for other tasks. We wanted to save time and costs.

It integrates well. We have SAP CRM and browser automations. We also have document automations. We also have API integration.

We have multiple bots running. It saves time. If manual invoice processing in SAP takes two to three hours on a daily basis, a bot can do the same job in half an hour. We are saving more than 70% of the time.

The feature that I currently like is Agentic AI process automation, where we use LLM models to automate our processes. I have heard a lot about it. I am eager to learn more about it and implement it in my organization.

It is very easy to use. You do not require any prior knowledge. Through Automation Anywhere University, you can learn it and implement your use case. A non-technical person can learn it within a month. For a technical person, it is easier to understand the loops, conditions, and workflows. A technical person needs only 10 to 15 days.

One improvement would be to implement a log mechanism within the workflow, similar to other automation tools. Currently, the log mechanism uses a text file. If we want to monitor a process, we need to check manually.

Another improvement would be to facilitate copying code blocks between projects, as currently, it is not possible to copy and paste them either locally or on the web.

I have more than six years of experience as an RPA developer and have used other tools. I have been using Automation Anywhere for the past two years.

It is very stable. I would rate it a ten out of ten for stability.

It is scalable. I would rate it ten out of ten for scalability.

We have about 1,700 automation users. We are a big organization with multiple locations globally. We develop automation for multiple departments and multiple entities.

Their technical support is good. I would rate it an eight out of ten.

Positive

I have used UiPath and Automation Anywhere. Both are very good. Automation Anywhere is cost-effective compared to UiPath.

The deployment is fine. It is very easy. 

It is completely on the web. We have our own licensing management tool. Anyone who has done internal training can request the license. A dedicated license is generated for you for a few hours or if needed, permanently. We have a dedicated portal for deploying our projects.

The upgrades are also easy. It does not require much maintenance.

We have multiple communities for automation. We have citizen developers and we also have expert developers. To identify the use case, we approach the entities and ask them to identify any mundane, repetitive, or complex task. We have a perfect pipeline for automation use cases. When a use case enters the feasibility check, an expert assesses its feasibility based on multiple parameters. If feasible, we proceed with development, prepare documentation, and design all solution documents. Next, we prepare the architecture, and development commences. After development, we conduct several rounds of UAT. Before deployment, we conduct a pre-delivery check with a dedicated department, ensuring best practices and proper exception and error handling. Once approved, we proceed with deployment.

In our organization, we are the first ones to work closely with the Gen AI and AI topics. We have a few opportunities to use the Gen AI capabilities in automation.

Overall, I would rate Automation Anywhere a nine out of ten.

Our primary functionality is built around screen scraping applications. We collect data from one application and either generate reports or use it to complete a user transaction or a customer transaction. For example, I have automation that handles bill pay. A request comes in, we verify it, and then we execute a bill payment transaction.

When we first started, the motivation was not wanting to be left behind in the latest and greatest opportunity for automating desktop applications that are boring and time-consuming but are still very important for the company's business operations.

We were able to see its benefits immediately during our proof of concept five years ago. We were able to take five users of daily work and reduce it to one user of support work. It was very significant.

Prior to implementation, these processes were done by the business units manually. Post implementation, the solutions are IT-based. As a result, IT now has additional responsibility for the business automation, just supporting the business function. It is one thing to build the software. It is another thing to maintain it and keep it running in production. There is a pretty significant lift or burden on the IT side that was not there during the manual executions.

I like the diversity. Every year, maybe even every six months, new modules are introduced to the program, and module enhancements are added to the toolbox to make automating various applications or situations easier. For example, the email module has been enhanced over the years to support all the latest authentication technologies. That is very important as we move away from username and password and embrace multi-factor authentication. Without the ability for these modules to stay in sync and up to date, we would not be able to use them. 

On the document automation side, the document learning features are not as robust as one would expect them to be. When I am doing document automation and a document fails due to low confidence in populating a field, it requires manual correction. Automation Anywhere states that if I correct enough documents, over time, the automation tools will learn where those error situations are and automatically fix them. We have not seen that feature work as advertised.

I have been using Automation Anywhere for roughly five years.

We have not had any issues. I am very impressed with its availability.

It scales quite well. I have not found an application that has exhausted the capacity of the system to execute. If we need to handle more transactions, we simply need to add additional bot runners. We currently have three bot runners in production. If we needed five or ten, I could have five or ten in a couple of days.

On a scale of one to ten, I would rank their technical support group a nine.

Positive

We have two other solutions in-house other than Automation Anywhere. They are very specific to the tasks they are handling.

It was pretty easy. I took a proof of concept that took nine months to build. Since I could not copy it from the proof of concept environment to our brand-new tenant or control room, we had to reprogram it. The reprogramming and testing took three months. Now, if you put all those numbers together, you probably could say it took us a year to build that first application.

It requires operation execution maintenance by my company and my IT group, but it does not take any Automation Anywhere resources to maintain it now. They maintain it at the cloud level and do a good job.

Two people were involved in the implementation.

The pricing is fair. My company is part of a big organization, so the pricing is very advantageous.

Five years ago, we evaluated four products, and Automation Anywhere emerged as the top choice. It offered the best cloud solution compared to the other vendors or providers. Now, if I took those same four vendors and evaluated them today, I might come up with a different answer. The industry has changed drastically just in the last three or four years, and if we did a reevaluation and could start over, chances are everyone, including Automation Anywhere, is on much better footing today than they were five years ago.

In terms of our AI approach, we are very cautious with AI. We are a Microsoft shop, so we are limiting our AI exposure to an Azure tenant-based AI resource. We are probably building our own LLMs to manage the intelligence or the exposure of intelligence to the AI engine. I do not see us ever using external AI modules.

On a scale of one to ten, I would rate Automation Anywhere a nine. 

I have worked on automation for several clients and organizations. In my previous organization, we automated several business processes related to governance. Currently, I am working with Nokia. Here at Nokia, we are performing data reconciliations, including financial data, procurement data, and asset-related data. To achieve this, we fetch data from systems like SAP and custom applications. Instead of manually navigating multiple screens in SAP, we use the RPA Automation tool to handle these activities. We are essentially automating these processes.

We are exploring an automated system that can help our production support system resolve tickets more efficiently. We have the data of or the history of multiple incidents raised in the past two or three years. We would like to utilize this data to resolve new tickets and provide the steps for resolution. When a ticket is there, the automated system can identify the solutions and automatically implement those suggestions to fix the ticket. We have not implemented it yet.

At the moment, we do not have any use cases for Generative AI, but we are looking into it. 

Automation Anywhere has helped us with efficiency and accuracy. We cannot achieve the same level of efficiency and accuracy through manual tasks. There are also cost savings. Manually performed tasks are financially demanding.

Automation Anywhere is easy to learn. People who are not technical can also get started with this tool. They can explore and slowly adapt to the tool. They do not need too much expertise, but having technical expertise can be an advantage. Overall, Automation Anywhere is fairly good for non-technical people.

We have integrated SAP and web applications with Automation Anywhere. We have also automated using the APIs. Its integration is good and has improved over time. It is evolving and getting easier for us to integrate with external applications.

We have had a lot of time savings. We have 40 to 50 bots.

The best aspect is the user interface. It is very user-friendly and snappy, responding quickly compared to other available tools. It is easy to learn, and the community support is good for asking questions and getting answers. Training camps are provided by Automation Anywhere. I recently attended the document automation training camp. In 2025, we are going to attend the AI agent training camp. The community and the training around the tool have been beneficial.

In Automation Anywhere, the logging process is manual. We have to manually log each instance where the bot fails or the progress of a task. Automating this through a new feature would help developers simplify their development process. That is my suggestion.

About a year or two ago, debugging was challenging in Automation Anywhere. If you had multiple task bots with a parent task and a child task, there were issues during debugging. Automation Anywhere took the feedback and fixed it in the next releases.

I have been using Automation Anywhere for over five years.

Most of the time, I do not encounter problems with the control room. On a scale of one to ten, I would rate it a nine, but I have had issues with IQ Bot. I would rate IQ Bot's stability an eight out of ten due to some service instability. For example, it sometimes fails to communicate with RabbitMQ. Automation Anywhere's control room is very stable.

It is scalable. Scalability also depends on how we code or develop the bots. We have 45 different use cases, and most of them are multi-bot, so I do not see any issues with the scalability. I would rate it a nine out of ten for scalability.

I would rate it an eight out of ten. Most of the time, when I raise cases for help with the Automation Anywhere vendor, I receive good help at the right time. However, sometimes I receive automated responses, which are not helpful and cause delays. Despite this, I generally get good support, so I can rate customer service an eight out of ten, but there is room for improvement.

Positive

I have used Blue Prism and Power Automate as well. In Blue Prism, logging is simple because it is automatic. Automation Anywhere is more snappy and faster in response. It also has inbuilt document automation, whereas Blue Prism does not have the inbuilt OCR capability. Overall, Automation Anywhere is a better solution than Blue Prism.

Power Automate is a very slow tool. If you have to open any flow, it takes a lot of time to load. This issue is not there in Automation Anywhere. It is more capable than Power Automate.

Currently, we have two control rooms. One is an on-prem control room, and the other one is a SaaS or cloud control room. We are using both.

The SaaS option is evidently easier because we do not need to manage servers, issues, or restart services. We prefer SaaS for these reasons.

In terms of maintenance, the platform upgrade from one version to another version is not difficult. We started with version 21 of A360, and we are currently on version 35. Most of the updates in the SaaS environment are automatic and happen during the night hours. Most of the time, we have not seen any issues. They went smoothly, but sometimes, there were some terminations and other issues. If those issues are also taken care of, it will be very smooth.

For the SaaS control room, there is not much maintenance. We are doing maintenance only in terms of the code. In terms of platform, we do not need any maintenance for the SaaS control room. For the on-prem control room, there is a lot of maintenance. Sometimes, the services go down, or the communication between the application and database server has problems.

We have seven to eight developers and three to four production support resources. Building a bot is not an easy process. It is a lengthy process. The duration depends on the use case. It can take from one month to three to four months to develop one automation.

We have a separate platform team that helps us in terms of the platform issues. We have a development team, and we have a production support team. So everyone has their own responsibilities and roles. 

We have about 45 use cases deployed in production.

I would definitely recommend Automation Anywhere. I have been using it for the last five years, and it is one of the leading RPA tools in the market.

Their community support is excellent. If we encounter challenges, we can rely on the community for help. There are ample training programs and an Automation Anywhere university to facilitate learning. The solution is reliable.

I would rate Automation Anywhere a nine out of ten. Some features, such as automatic logging, could be improved. It has the potential to become the best tool.

I started with Automation Anywhere version 11, which was the desktop-based version. It was the most initial cloud version. There is also a 2019 cloud version of Automation Anywhere. With some modifications and improvements, 2019 led to A360, which was again a cloud version but an improved form of the 2019 version. A360 is the current and running version of Automation Anywhere.

Most of the problems we've worked to solve have been related to business process automation. We were able to automate email - including replying to email, and capturing data.

We've also used it for website automation. We can do object cloning in PRA terms.

We were able to automate several data systems. We've automated Excel, for example, and various backbones of process systems.

It releases humans from monotonous tasks. Automating helps a lot and gives us the ability to free people for more high-end activities.

Most of the problems we have solved are related to business process automation. Ideally, in RPA, around 80% of the problems are related to business process automation. 

Processes and tasks automated compared to a manual approach offers a lot of benefits. Any RPA tool will offer good time savings. When we have to do things manually, if there is a 20 or 30-step process, as a human, we may get confused and redraw information. We make mistakes. All of these things take time. Things can be very data-intensive and this takes time for a human. They need to look it up and validate, et cetera. and there are high chances of error. A human can misread or misunderstand, this could lead to losses. With RPA, we gain time savings, and we can make everything 100% error-free.

To train a non-technical employee, it might take a bit. They need to learn the basic foundations. It might take a month or two if they give time to learning every day.

It helps increase productivity by 20% to 30%.

I can integrate tools with Automation Anywhere. We've done Excel and database automation, web applications, data tables, FTTP tables, et cetera. 

Copilot is an area with a lot of scope for improvement. It's the number one thing to improve. It just gives us a structure. It doesn't automate the workflow. This is a major area of improvement. Currently, it's still evolving. The integrations in Automation Anywhere are evolving, and going forward, they could be more beneficial for end-to-end processes. It should not be just an add-on, but it should be very well integrated.

I have been using it for close to four to four and a half years now.

The stability is okay. I would rate it 8.5 out of ten.

I would rate the solution 8.5 out of ten. It is quite scalable. 

Power Automate is quite good, and gaining ground. With Automation Anywhere, you can log into the control room and work on it.

The deployment is on the cloud. The process depends on how straightforward it is. It depends on the number of steps. The deployment time doesn't take too long, however, they could improve a bit on the scheduling. It could be a bit more user-friendly if there is a new person coming in and using the solution for the first time.

Deployments tend to take a few days once all the approvals are in place.

The solution does require maintenance. Once you have developed and deployed a bot, there could be many things that may break the bot. We need to maintain them to keep them running as expected.

The ROI may be high. It depends on the process and the number of people involved.

It's not cheap. It's expensive. Power Automate is much cheaper.