We would like to use the IQ Bot only, for PDF extraction. Currently, we are using PDF data extraction and finding whether the data is present. If we go with the IQ Bots, the main reason will be the capability. At the moment we are having some sort of problem with reading the images. We are using OCR and Python for that right now.
Robotic Process Automation (RPA) Python Reviews
Showing reviews of the top ranking products in Robotic Process Automation (RPA), containing the term Python
Automation Anywhere (AA): Python
When it comes to integrating the solution with other applications, there are some challenges. For some third-party solutions, there are no direct interconnections. For example, there were no direct connections with SAP systems. So, we had to create connectivity between Automation Anywhere and some third-party solutions. They have now improved that situation a lot and we can connect SAP and other systems as well.
If they want to sustain their position in the market, they have to be flexible, working on how we can integrate with third-parties, working on a machine-learning product. People are expecting that and it would be really helpful.
From the IQ Bot perspective, frankly speaking, they still have to improve a lot. I got IQ Bot training in San Jose. My expectation from a straight, technical, architectural point of view was that I would be able to create my own algorithm and integrate it. But with IQ Bot, I am not able to integrate anything. It is already well-defined, so I have to use that particular option only. I know I can not go with any other machine-learning platform. I hope they will be coming out with version 12 where we can integrate it with Python algorithms and other stuff. It might only be in the future, it might only be on the roadmap. But as of now, it is lacking a lot in that area. We are expecting, for most of the documentation, things like tags, that there would be a checkbox option. That's lacking in IQ Bot.
It has been challenging integrating Automation Anywhere with other solutions, due to all of the controls, as we are a bank.
We are really excited to have Python run directly from the bots. This would be a massive upgrade, as we could integrate AI even further with the solution.
There is room for improvement in AI, but they are working on it.
They can work more on integrating the code. I would like to see more similar examples where we don't have to create the code in the bot, but can create the code in the Workbench.
A useful feature of Automation Anywhere, which is on the way, is the ability to use inline code for both Python and VBScript. The reason that this is so useful is a lot of the time people who are doing RPA are people with technical backgrounds who have been developers. Although, Automation Anywhere is very powerful, sometimes what you need to do is a very specific task which is accomplished more efficiently with code. So, I'm really excited that this feature is coming soon.
Another feature that would be cool to see in Automation Anywhere is more use of artificial intelligence. Right now, RPA is basically limited to a bot which is simply following the instructions that it was given. However, what would be really useful is if a bot was able to more flexibly respond to issues. For example, sometimes there are erroneous errors and pop ups. With good code, you can certainly build your bots to be able to handle that. However, it does take time and a bit of technical know-how to be able to do that. But because a lot of those technical challenges tend to be pretty similar, I would like to see Automation Anywhere include more robustness into the way that errors are handled.
The biggest area of improvement with Automation Anywhere is on the enablement side. A lot of the times, people have the capabilities with the tool, and it's all there, but it's not necessarily so easy to see that roadmap to RPA success. This is very new technology, and there's not a pre-walked path to success. It's unique to every organization, but there are certain commonalities. What Automation Anywhere could do to improve the solution is help customers understand what the tool is really capable of. Walk customers through what changes are needed both organizationally and in terms of infrastructure to really see success with RPA.
I've seen customers who have the licenses. They have the know-how. They have all the pieces from Automation Anywhere to achieve success. But, where they fell short was there maybe wasn't necessarily commitment from the right stakeholders within the organization or maybe there were other forces which were holding things back. What they needed was vision for it:
- How can they operate efficiently?
- How should it work?
- How should it look?
- What roles are needed?
These are the areas where I've seen customers struggle, and somewhere where Automation Anywhere could really improve.
I think there are quite a few new features which have been introduced, but right now we've been using a lot of object cloning and screen scraping. Those are some of the primary features, but we've been able to use other features very wisely as well. These include inbuilt email functionality, some integrations to incorporate heterogeneous systems like Python scripting, and also other things.
Compared to the rest of the industry, this product is easy to use, not only from the technology perspective but from the business user side as well. We have been very successful, in that we haven't seen a situation where we trained a particular set of users or customers, and they complained that they are not able to use it.
Some of the other products in this industry have not been able to succeed because of this issue. Although they are really great, they are too focused on technology and IT, rather than business focused.
Automation Anywhere needs to improve Excel and its commands. In Excel, if you want to add a sheet, you cannot add a sheet. There is no command for adding a sheet. Also, if you want to rename something in Excel, you cannot rename it. You always must keep a list that you can rename through your keyboard, but there is no specific command in Automation Anywhere.
The email automation command also needs to be improved. The email automation command sometimes is not working properly. When you connect the mail server, it is not being done properly.
The OCR feature is not that reliable.
With the PDF command, you can only read structured data. If your data is in an unstructured format, there is no command for it in Automation Anywhere. Then, people need to use Python coding if their data is in unstructured format. Therefore, Automation Anywhere needs to improve in the PDF area.
We have had issues with stability in the past, and there should be a function that helps you to find out what is wrong with your environment that is causing the automation to fail.
We would like to have more information on how Python integration works, and how easily it can be done.
We have a situation where the system is not capturing data properly. It is clicking on the UI but it is not able to expand.
We have trouble with large volumes of data in Excel and are not getting good results. I want all the functionalities that they have in the Excel added to our existing version or an upcoming version of the product, e.g., Macros and VLOOKUP.
We have had connectivity issues, such as the reporting of a bad gateway.
We have integrated Automation Anywhere with our SQL database. I heard in version A2019 that we can easily integrate Python too, which is great news. Up until now, I have not been able to implement Python in our Automation Anywhere task bots. Though, we can integrate with other applications, like artificial intelligence, and this is better.
We have not implemented this tool with our AI. We have AI and IBM Watson. We would like to get information (or services) from the Automation Anywhere team on how to connect and implement the AI, as this is a great thing in the market. I'm looking for what to do.
Sometimes, we are facing issue in unattended. If unattended mode is not working, sometimes it will work in attended mode. However, I don't want anything in attended mode. The process is rule-based and logic-based. We just have to schedule, then if the user wants to run from the Control Room, they can. To run attended mode, we have to login into the VM to start it and human activity is required.
We would like to be able to directly implement automation for Word documents and XML. Currently, we have to write a script or some code in Python, and there is no direct solution. Basically, I'm expecting some things that are more innovative.
It would be great if technical support were a little faster.
I would like to see more documentation for automation.
The GUI can be improved by making it more user-friendly.
Python integration is coming with the next release, and they should expand on that.
We have automated web scripting, downloading documents from particular websites, scraping data from websites, and extracting the data from PDF files. We are also updating the application with some user inputs. E.g., if you have an Excel sheet to update into a particular application, we can use Automation Anywhere tool and automate the process.
We have built some Java applications in Automation Anywhere. I have also coded macros and Python scripts to include in our Automation Anywhere tool.
The primary use case for this solution ranges from the web scraping to working on more complicated applications, which are built in-house. For example, in the case where we would have to get into an application that is not automation-friendly, we would use the OCRs to recognize the fields, insert information into the application then extract some of the information. We also do a little on the database front.
We are a solution provider for a back-end team who works on a different manual task.
We tried the same type of flow automation using a Python and VB application, but we don't have the leverage of OCR which is available from Automation Anywhere.
Most of the automation that we currently handle is that which has already failed when using the previous versions with the VBAs and Python.
We are using this product on-premise, so we still haven't upgrade to the cloud.
The product is not working sometimes with the methods we prefer. If it is going to be a flexible product it should include some more common things. There are so many small issues with the product it is not as good as it could be.
The most valuable features of this solution are its user-friendliness and the flexibility in terms of programming language. We can use any programming language, and I have just started to learn Python. I want to next look at real-time scenarios and see how Python can be used to help write bots. Having no restriction for any single programming language is very useful.
This solution uses Artificial Intelligence and Machine Learning to help automate tasks. Our analytics people and data scientists are looking forward to seeing how this will help.
We are looking forward to having the capability of building the IQ Bots the way we want, using Python scripts. With our scripting, we can fill in the gaps and improve accuracy.
My advice to anybody who is researching RPA solutions is to consider several things. I can see that the path that Automation Anywhere has right now is going to take it to the highest levels. My suggestion is that if you have a lot of coding or scripting experience then I would recommend UiPath. It has its own plusses and minuses, as with every tool.
Instead, if you're looking for someone that can you you build a bot within a short time, who doesn't know any coding or have a background in programming, then I recommend Automation Anywhere. You cut out the cost of developers, but you add to your cost in terms of licenses.
Also, for manageability, Automation Anywhere is the best choice.
I would rate this solution a seven out of ten.
Alert transition is happening in the Python space and on the front lines of AI and with other tools. You have the opportunity to use different data blending tools and data manipulation tools. It would be good if Automation Anywhere could make more plugins available to use with a greater variety of tools and additional options. I think it would be a better product if it had a greater capability for integration.
Go through the Automation Anywhere courses properly to be aware of all the system variables and all the features it has. Then proceed with the development phase. Until you're clear on the basic things you cannot do anything in the tool. You need to know those things.
I have done the Masters course on Automation Anywhere University. I have a Masters Certification from them. Their courses are pretty interesting. At the end of the Masters course, we had an interview on the course itself. There were a lot of informative things and it was a great experience.
Next year, we should be moving into cognitive document processing.
I have personally used RPA before. It is just the buzzword now.
At the time, we were looking at or looking at replacing our complex, legacy platform which had a lot of steps. Our multiple, previous software tools that had a lot of Python, Perl-based scripting, and Perl website scrapping plugins.
MetaBot still needs to be optimized, though Version 11.3 did integrate all the commands into it. We are not able to see all the lines of code which are executed in MetaBot.
Integrate .NET functionalities instead of installing Visual Studio, which is an additional resource to us.
Needs more Excel features and functionality.
We are moving onto Python machine learning and would like to have additional functionalities to be added, especially to MetaBot. This would help for the use cases that we are working on.
There are some discrepancies in respect to the variable creation of copying and pasting the code. However, in version 11.3, these type of disturbances has been removed.
We work in a very complex type of environment. We do a lot of data analytics and strategy consulting work. The work that we have automated so far is our legacy suite of applications, doing all the reporting, data extraction from different web sources, and collating the information, then publishing to websites.
We use RPA or the Automation Anywhere tool to orchestrate the whole process. We combine it and integrate it with other API and Python solutions to help get the data from a particular source, whether it be an FTP source or AWS environment. Then, we make those validations using Excel. After that, we do our QCs and validations, then publish or submit the reports.
We partner with a few other vendors as well for different use cases based on the type of clients that we are working with.
I would like to leverage the AI component without the requirement to know a coding language, such as Python.
I would like Automation Anywhere to integrate with Google activities, like Google Sheets. While this should be available in A2019, I would like it to be scaled more.
Version A2019 will have better error handling. They are also adding Python script.
I would like them to add dictionaries to it.
There are multiple features, like PDF integrations, which are coming up. Task-related features to PDF integrations are coming up nowadays, so people want those executed and improved a lot in Automation Anywhere.
For improvement, I would like to see Automation Anywhere integrating with multiple other technologies. As of now, it is integrating with .NET. When it come to future technologies, I want Automation Anywhere to integrate with Python scripts, and make the execution easier. That will be very helpful, having our cognitive technologies interacting with Automation Anywhere.
I can introduce IQ Bot. I can introduce my Python script that I'm trying to work on. If I'm not able to make it with an Automation Anywhere, I can go out-of-the-box. I can try to build in my own programming language where I have expertise or I might not, but in this way, I am putting it across different scripting mechanism and running it the same bot. This is something which I think is very good and we can scale again to a different level with different use cases.
We have our limitations. We can run only VBScripts and one more thing. If we want to run Python scripts or batch files, we cannot do it. So, if they would add these extra things in there then that would be beneficial.
I have to add in Excel commands because it is a limited feature, as we only have five to six commands. Suppose we have a spreadsheet and we have to get the counts for how many rows were filled, we have to do it through a Macro or VB. We have to write a separate code and call through Automation Anywhere. If they added in this inbuilt feature into Automation Anywhere that would be helpful for us.
When we started Automation Anywhere, it was a rule-based automation. All the rule-based used cases we could automate. Now, with the upcoming version, it includes AI and Python, so we can go ahead in our roadmap and start automating dynamic used cases too.
Whatever I was thinking for version 11.3, it's being implemented in the A2019. For example, it is great that the product will be:
- Integrating with cognitive analytics.
- Integrating with Azure.
- I can write scripts.
In version 11.3, if I want to write a Python script, then I write an ESX file. This is gone in version A2019.
Stability is good. It is improving. Even with the newest version, they have improved a lot. Earlier, we used to face some blockages. Now, with the newest version launch, there will be so many new functions.
Previously, there were very few functions for Excel automation. We had to put in logic for everything, like in other coding languages. Now, they have included so many functions in Excel. You can easily automate your Excel work to the function without writing a script. Hardly, a script is required.
We have the option of handling exceptions. They have given us so many options to handle exceptions. If your bot crashes, you are trying to find some page, or you're trying to find some file and it is not found, we have alternatives. We can handle the exception, like coding in Java or Python. We can handle the error because we have the functionalities. Also, we can write a small code for handling all these exceptions. So, it is easy to use without any fuss.
- The universal recorder is a kind of hybrid recorder that enables the user to record the data across devices.
- Option to send email from Outlook or SMTP.
- A2019 contains almost 11 diffs. datatypes. Ex: window,number,table.
- Browser support is Improved with Different Browsers support: IE, Mozilla, Chrome.
- A2019 is not dependent on single os. It supports Linux, Windows, etc.
- Running bot through Excel, Waiting for release from the vendor
- Instead of meta bots we can use run DLL commands
We have found both the basic Task Bot & IQ Bot very useful as the combination of both helped us to create some solutions which were End to End Business solutions. IQ bot performs the basic Data Sample set creation and Extraction of the details from the unstructured PDFs and later we have used the Business Rule engine using the IQ Bot to perform the data feed into the various Systems. Overall we were able to achieve the value using the combination. As part of our internal solution, we have also created custom Python scripting to handle the OCR failure issues surfaced due to IQ Bot product limitations
- Unlike other leading RPA tools, UiPath provides a great user interface so that a person with minimum technical knowledge can understand the flow.
- UiPath provides a large number of activities which we can reuse as per our need just by changing the properties and parameters.
- UiPath provides the flexibility to create custom activities as per our need.
- UiPath supports integration with various OCR, cognitive services, ML and IoT related services and tools.
- UiPath provides the flexibility to use an unattended robot without the need for Orchestrator.
- UiPath Orchestrator component is a very powerful tool which if utilized effectively, can make the robot management and monitoring very easy.
- UiPath Orchestrator supports using APIs, in which we can get the details from the Orchestrator database.
- UiPath’s recording feature is very helpful to begin the automation process and provides options to choose from recorders as per our need.
- UiPath supports version controlling feature which is very useful when the project is done by sharing tasks with multiple persons in the team.
- UiPath provides various workflow models such as sequence, flowchart, and state machine.
- UiPath provides an inbuilt logging facility.
- UiPath provides scraping of both structured and unstructured data.
- UiPath provides specific tutorial and documentation for SAP automation.
- UiPath provides debugging options usage, in which we can go through each activity along with the process flow, highlighting each action on the screen along with the flow, and showing the values of variables at the run time during the point execution of an activity.
- UiPath provides the facility to use numerous .NET namespaces by importing them easily to the project.
- UiPath provides the flexibility to create a custom activity from our custom .NET DLL.
As a software engineer, the features I felt the most valuable are given below:
- Version control
- Python support
- The wide knowledge base on the internet.
- The UiPath forum is awesome; we can find solutions to every problem from there.
- It has nice support too.
I would like to see some sort of better implementation of the Python modules. Right now, any error code that Python throws up turns into a generic error that is difficult to track down. I would want to see some sort of integration that at least lets me debug the issue.
Its pretty difficult to do source control through the typical Git functions because of the nature of the UiPath and the visual element. Some sort of method to bring source control home for us would be very helpful. A way to systematically track which actions or activities were added and subtracted from a given process before adding to a repository would be very helpful to us right now. Our team of developers is growing, so control has become somewhat of an issue for us.
We have used this product to implement automation for several things, including:
- UI automation that updates a project's status based on conditions
- Gathering data from web sites
- Excel functions related to HR operations
- File checks, to determine the validity
- Python and .NET scripts to connect to different databases
In the majority of cases, the initial setup of this solution is simple to medium in terms of complexity. We are finding very few complex scenarios at the moment.
I think the overall architecture is simple. It is very clear and very straightforward. UiPath's product team is doing a great job in is creating a lot of very out of the box integrations and analytics, and that always helps. That is good, but I think if people are not trained yet and they think that it's easy, drag-and-drop, and simplistic, those folks struggle a lot.
We've seen that people think "Oh yeah, it's just some scripts and drag and drop so we can do this easily" and that misconception exists. We don't treat it as an easy scenario, so we gave it all the respect that proper Python code, a data science problem, or a highly complex situation deserves. When you approach it that way, it's at best at a medium complexity.
In general, we treat it right in the middle. It's not that straightforward, but the architecture is simple enough that the development complexity is medium. That's the simple and medium combination.
I found the Library option as most valuable because first, we can avoid the development of the same task again by developing and publishing a task as an activity into the Orchestrator, and second, in case of any changes required, the developer can make the required changes and republish in the Orchestrator with the new version. At this point, others can download that task as an activity and can use it with their Bots.
Apart from the library, I like the API activities and the activities which we can call Python scripts from.
Orchestrator is the feature that allows making use of robots for remote location systems. The work queue is the added advantage to maintain the bot data.
The inbuilt RE Framework makes development easier and fast.
UiPath helped me to automate the various complex and mundane processes. You can also run your Python and .NET code in UiPath, which makes the development easy.
The UiPath interface is very user-friendly and also it is easy to learn and use.
There are many online certifications in the UiPath Academy, so it helps the user to become a certified developer and learn everything from the online courses.
Support from the UiPath community is very good.
Following are the most valuable features of UiPath:
- Every month, there are new features released. These are all next-level and can disrupt the automation industry.
- Vision for RPA and ML work together including ML skills, ML extractor, and Python activities.
- UiPath is easy to use with very simple drag and drop functionality that allows you to automate processes without knowledge of coding.
- Covering automation in all possible applications including SAP, G Suite, Salesforce, etc.
The product needs to be "simple and easy" to use.
A good example of this is the use of Python code to perform automations. It is free and the skills transfer is able to happen quickly and the operation and initiation of the completed automation are easy to execute.
The primary focus of the Robotics Process Automation software should be to enable business users to perform "codeless" automations that will still fall within the strict governance and compliance structures within which we are required to operate.
While we have got WorkFusion as a solution at the moment, we continue going through due diligence. We actively check and evaluate the alternatives that are available. We consider it an ongoing or proactive review of our current toolkit. We currently use mostly WorkFusion, but we also use Python for some of the other automation. We are looking at other important products like UiPath, Blue Prism, Automation Anywhere, and Help Automate. Those are the tools that we are pulling in to use for our comparisons. One of them may solve potential pain points the current solutions do not.
SAP Contextor: Python
With Contextor, it is possible to add functionality and add capacity with another language, much like Python or C#.
The older models of Contextor are good and competent, with good skill, good practice, and the level of precision is important.
One of the biggest differences between Contextor and other solutions is that it is not based on the OCR functionality to discover certain applications like Citrix or remote access.
Contextor is a good solution for a Citrix environment. It's the first with this tool. Other tools provide the same functionality, but not with the Citrix environment.