It's difficult to say what I'd improve in Tungsten TotalAgility because we haven't been actively marketing that particular product in the last probably six to nine months. The speed of product development means that my experience with it is likely out of date in terms of the product suite capabilities. The difficulty with customers nowadays is that they are subject to the winds of change in terms of the adoption of cloud technologies and the offerings those cloud technologies bring. Microsoft, for instance, is rapidly filling in the spaces between various product areas that other vendors typically owned. That makes it difficult for customers to settle on their functional requirements because the offers are competing in the same space. I look for more integration with mainstream subscription based platforms, more security integration, and possibly some more lifecycle management features. However, the difficulty with that is the competition with existing options that people have in their subscription library.
Software Engineer at a consultancy with 11-50 employees
Real User
Nov 10, 2022
Kofax TotalAgility could improve the OCR engine. The page OCR engine is not accurate in predicting the data properly. And also the Trainable document separation methodology is not accurate in classification even with a significant larger data set.
Javascript and jQuery integration should be increased because some times we have to create custom solutions in other languages as C#, asp.net and then integration.
advance SQL query tool should be there so that user could right query directly there and should able to preview the results. Existing query configuration is not efficient.
Director at a tech services company with 11-50 employees
Real User
Sep 8, 2022
The pricing for Kofax TotalAgility could be more attractive, particularly the cloud version of the solution which was offered to my company. This is its area for improvement.
In terms of OCR, if you see the OCR currently is using some omni reader, such as ABBYY OCR engine. There is a lot of OCR technology. It is all really good now. If you look at the OCR by Google or the other Microsoft OCR we have, we get a good OCR result compared to the Kofax OCR engine. There should be some integration allowed so that we can use the OCR results from any other good OCR engine. While the image, the field extraction, is really good in KTA, the OCR result is not so good. The handwritten aspect is not so perfect. It’s still maybe 40% or 50% accuracy. It is not easy to read handwritten text since the handwriting will differ from one person to another person. It needs improvement. When you look at the Google OCR engine, the OCR result of your handwritten data is not so good. There is something called Quick Capture. Quick Capture is something that is helpful for the end user. With it, there will be less dependency on the developer. However, we can't do table line item level extraction. Table line item extraction is not possible through Quick Capture. We’d like to have this capability.
When I'm talking about low-code development, they are consulting on a new product with low-code. This means that now you won't have to build that script. It's just a simple single centralized development tool where you can drag and drop things. You can just drag and create your form using drag and drop features, and you can implement your own business logic. It’s not truly low code yet, as every two or three projects, you will have a situation where you have to go behind the essential things. You will definitely still need to customize. You cannot yet really rely on low-code development. If you have a complex business requirement, you must build that complex script. Previously, it did not support all browsers. However, it’s my understanding that they have since changed that shortcoming. They have to create a few different packages. For example, let's say we are in the insurance sector. The insurance sector will have a specific set of processes. If they could give us a framework where I don't have to be concerned about how to build the form, something already heard towards insurance, that would be ideal. At least 80% of my time will be on implementing business cases rather than building software processes. I just want this to have some use case differences and some pre-built solutions based on industry. That way, a user can directly apply their business cases to that and it'll be like a ready-made solution - or at least 90% ready-made. The last 10% we can feed our inputs into more easily.
There needs to be more accessible certification and better pricing for Kofax TotalAgility. Kofax launches a new version every year, and we experience some issues. We've worked on versions 7.2, 7.3, 7.5, and 7.9, and we are currently using the 7.10 Kofax Total Agility version. The solution also provides several extraction features, NES quality, and VRS profiles, giving the front-end soft utilities. A downside is that the tools are very costly and challenging to learn. You will need to purchase a Kofax License learning tool and certification to learn about the Kofax Total Agility. It is expensive, and if training could be provided on YouTube, there may be no need to get the certification and invest that much money. The learning tool shows you how the use the application and how you can improve your business by using the tool. There are limitations to the software. For example, it does not support G-Queries, you can write in only C Sharp, and there is support only on the Windows server. There are also limitations to the hardware, probably due to Kofax's policies. For example, because of the number of applications, you can only do LEMCAD and can't do LEMCAD designs for other forms.
ESM consultant/ Kofax developer at a tech services company with 11-50 employees
Consultant
Jan 17, 2022
Sometimes there isn't enough accuracy when it comes to extracts. A user has to validate the data to make sure that the data is correct. If that step is improved, then it's perfect. We need a level of hundred percent on the extracted data. Right now, we need a user to validate and make sure that the data is correct, which is an extra step and takes some time. This is especially true if the documents we are extracting from are handwritten. A's sometimes read as 8's, for example. For the most part, they have so many different aspects right in Kofax, I can't speak to if there are any missing features.
Tungsten TotalAgility revolutionizes customer experience and organizational efficiency by integrating content-intensive workflow automation with advanced technologies like low-code process design and AI-driven intelligent document processing. Offering multiple deployment options including public cloud or preferred cloud service, TotalAgility empowers organizations to choose the solution that best fits their needs. With leading document intelligence, low-code automation, and end-to-end...
It's difficult to say what I'd improve in Tungsten TotalAgility because we haven't been actively marketing that particular product in the last probably six to nine months. The speed of product development means that my experience with it is likely out of date in terms of the product suite capabilities. The difficulty with customers nowadays is that they are subject to the winds of change in terms of the adoption of cloud technologies and the offerings those cloud technologies bring. Microsoft, for instance, is rapidly filling in the spaces between various product areas that other vendors typically owned. That makes it difficult for customers to settle on their functional requirements because the offers are competing in the same space. I look for more integration with mainstream subscription based platforms, more security integration, and possibly some more lifecycle management features. However, the difficulty with that is the competition with existing options that people have in their subscription library.
The product's console version is old. It should also improve its forum.
It would be nice if the tool has more connectors to different systems. It is an expensive solution.
Kofax TotalAgility could improve the OCR engine. The page OCR engine is not accurate in predicting the data properly. And also the Trainable document separation methodology is not accurate in classification even with a significant larger data set.
Javascript and jQuery integration should be increased because some times we have to create custom solutions in other languages as C#, asp.net and then integration.
advance SQL query tool should be there so that user could right query directly there and should able to preview the results. Existing query configuration is not efficient.
The pricing for Kofax TotalAgility could be more attractive, particularly the cloud version of the solution which was offered to my company. This is its area for improvement.
1-Database level command execution time should be available.
2- Javascript and jquery integration should increase.
In terms of OCR, if you see the OCR currently is using some omni reader, such as ABBYY OCR engine. There is a lot of OCR technology. It is all really good now. If you look at the OCR by Google or the other Microsoft OCR we have, we get a good OCR result compared to the Kofax OCR engine. There should be some integration allowed so that we can use the OCR results from any other good OCR engine. While the image, the field extraction, is really good in KTA, the OCR result is not so good. The handwritten aspect is not so perfect. It’s still maybe 40% or 50% accuracy. It is not easy to read handwritten text since the handwriting will differ from one person to another person. It needs improvement. When you look at the Google OCR engine, the OCR result of your handwritten data is not so good. There is something called Quick Capture. Quick Capture is something that is helpful for the end user. With it, there will be less dependency on the developer. However, we can't do table line item level extraction. Table line item extraction is not possible through Quick Capture. We’d like to have this capability.
When I'm talking about low-code development, they are consulting on a new product with low-code. This means that now you won't have to build that script. It's just a simple single centralized development tool where you can drag and drop things. You can just drag and create your form using drag and drop features, and you can implement your own business logic. It’s not truly low code yet, as every two or three projects, you will have a situation where you have to go behind the essential things. You will definitely still need to customize. You cannot yet really rely on low-code development. If you have a complex business requirement, you must build that complex script. Previously, it did not support all browsers. However, it’s my understanding that they have since changed that shortcoming. They have to create a few different packages. For example, let's say we are in the insurance sector. The insurance sector will have a specific set of processes. If they could give us a framework where I don't have to be concerned about how to build the form, something already heard towards insurance, that would be ideal. At least 80% of my time will be on implementing business cases rather than building software processes. I just want this to have some use case differences and some pre-built solutions based on industry. That way, a user can directly apply their business cases to that and it'll be like a ready-made solution - or at least 90% ready-made. The last 10% we can feed our inputs into more easily.
There needs to be more accessible certification and better pricing for Kofax TotalAgility. Kofax launches a new version every year, and we experience some issues. We've worked on versions 7.2, 7.3, 7.5, and 7.9, and we are currently using the 7.10 Kofax Total Agility version. The solution also provides several extraction features, NES quality, and VRS profiles, giving the front-end soft utilities. A downside is that the tools are very costly and challenging to learn. You will need to purchase a Kofax License learning tool and certification to learn about the Kofax Total Agility. It is expensive, and if training could be provided on YouTube, there may be no need to get the certification and invest that much money. The learning tool shows you how the use the application and how you can improve your business by using the tool. There are limitations to the software. For example, it does not support G-Queries, you can write in only C Sharp, and there is support only on the Windows server. There are also limitations to the hardware, probably due to Kofax's policies. For example, because of the number of applications, you can only do LEMCAD and can't do LEMCAD designs for other forms.
Sometimes there isn't enough accuracy when it comes to extracts. A user has to validate the data to make sure that the data is correct. If that step is improved, then it's perfect. We need a level of hundred percent on the extracted data. Right now, we need a user to validate and make sure that the data is correct, which is an extra step and takes some time. This is especially true if the documents we are extracting from are handwritten. A's sometimes read as 8's, for example. For the most part, they have so many different aspects right in Kofax, I can't speak to if there are any missing features.