We are using the tool to capture data from bills that customers send in.
It is not used on the business side. It is on the back-end side, and it is sort of automated. We process things for the digital channel.
IBM Datacap helps you streamline the capture, recognition and classification of business documents and extract important information. Datacap supports multiple-channel capture by processing paper documents on scanners, mobile devices, multi-function peripherals and fax. It uses natural language processing, text analytics and machine learning technologies, like those in IBM Watson, to automatically identify, classify and extract content from unstructured or variable documents. The software can reduce labor and paper costs, deliver meaningful information and support faster decision making.
IBM Datacap is also known as Datacap.
Download the Intelligent Document Processing (IDP) Buyer's Guide including reviews and more. Updated: November 2021
Turkcell, PowerSouth Energy Cooperative, Central Nacional Unimed, Conqord Oil
We are using the tool to capture data from bills that customers send in.
It is not used on the business side. It is on the back-end side, and it is sort of automated. We process things for the digital channel.
We don't use it internally in the organization. We link into it and expose it to customers. Generally, it hasn't change the way we work, but we hope it will change the way our customers interact with us.
The most valuable feature is its ability to capture data, which changes all the time into different formats.
The usability is great. The tool is powerful, and it's the right thing for us to use.
The integration process was pretty easy. There were exposed APIs for us to call and the documentation was available for us. Therefore, we could do it pretty easily.
I would like to see the product have the ability to process more documents in parallel. Right now, it is a single queue. Therefore, if you want to really test the load and stress test it, having multiple instances and the ability to scale it up would be great.
The stability is great.
I would like it to be able to handle multiple instances. E.g., if we had suddenly one million documents to load into it, the solution would work through them one by one. It would be great if it could dynamically scale up and have three or four instances that it could handle, or even many, concurrently.
I have a direct line to them, which is amazing. They are very good.
We were not using anything previously. This was something from our innovation stream. It was a new experiment that we wanted to test.
The initial setup was easy for us.
We worked directly with IBM.
It is supposed to reduce costs.
Pricing depends on how much we use it. We pay per bulk quantity. We pay as you go. Therefore, it sort of depends on our usage of it.
IBM offers a strong product. We also looked at Google's offerings with something similar. There was another company that we looked at, which I can't remember.
What differentiated IBM was we had a strong relationship with them already. It was a natural fit for development and support throughout our process and journey of this application.
It is not easy, but it is worth the time to configure and set up. The benefits dramatically outweigh the cost and labor of implementing and using it.
Make sure that you are using it for the right reason. There has to be a compelling, valuable reason why and where you are going to put this product in, and you have to pick the right thing. Otherwise, you are wasting time and money.
We are using it for automation projects.
It is something that we want to continue to invest in and use.
Most of the algorithms in IBM Datacap are for capturing information on physical, tangible pieces of paper or documents, allowing them to be scanned, and associating them with an automated workflow.
Datacap can work in conjunction with FileNet most of the time. However, Datacap allows companies or government municipalities to take in information, scan it, and have it in a centralized database, so you can have an automated workflow structure for it. It speeds up a lot of internal processes, reduces human error, etc.
We use a lot of in-house ECM solutions and a lot of different technology solutions. We have our own CRM system. As far as Datacap and FileNet go, we scan in a lot of our documentation and have a lot of automated workflow structures for building RFPs in the government sector. Therefore, we can take a lot of information and scan it in, then it can tell us the differentiation in documents, etc. We try to be as paperless as possible. It's mostly digital these days.
I'm working with the Connecticut State Troopers right now. We're implementing a Datacap and FileNet system for them. Their goal is to have their sex offender registry unit:
Thus, we have integrated a system, which should be going live in the next few weeks for them. The Datacap and FileNet systems will work in tandem together to pull the information from the letters, mailing, and forms which have been sent out for the sex offenders into the system. Everything will be an automated workflow structure so they don't have to hire ten people to sit there and type in all the information or scan the information in and still have to type it. It should automatically pick up a lot of the details which are already on the documents, then be able to flag specific individuals, and tell the people who are working at the state trooper facility, "We need to check up on this person. This person moved."
This is one specific example that I could give you as far as the benefits go. It's a lot of time and money saved, and ultimately, it protects more people, because there is less human error. If someone is typing in the information and they screw up, it could be detrimental and somebody could end up getting hurt in the process. So, there are a lot of different perks to it.
Automation is all about eliminating human error and saving time. Time is a way more valuable resource than money. You can take out loans and receive money from anywhere. Thus, time is a much more valuable resource.
The overall issue is the amount of money that you pay workers every year to spend time filing paperwork or sticking with a manual process. An example: In the state of Connecticut, we implemented a solution which was forms based, but inspired by IBM technology. We were able to eliminate a process which normally took 60 days and cut it down to six hours.
I would like to see integration of Watson AI technology into Datacap. AI could play a huge, pivotal role as far as where IBM technology's heading. Not only that, but where the world's technology is heading, as well.
We have been seeing all these different competitive uses and different, softer technologies along with a lot of copycats who have popped up. IBM is fantastic and phenomenal at being a differentiator. As time has moved forward, there has been less focus on the Datacap and the FileNet side, and there has been more focus on the next generation of technology, like AI.
IBM works in both the private and public sectors.
As you're implementing the system correctly and you have the right people able to run and support the system, you will rarely ever see issues. When you compare that to a manual process, it's uncanny the amount of benefits that you can get from such a stable system.
It is very scalable.
The biggest thing that IBM has going for it on the customer service side is their ability that they have to create relationships.
My company adopted the technology in the mid-2000s because they wanted to make things faster, better, stronger, and more efficient. They also wanted to stay ahead of the competition and kept information secure.
It can take some time to implement. As long as the customer understands that they have to be patient with the entire process, it's very easy to work with people and have the level of the technology they want to work. However, everybody has to be onboard with it.
It is wonderful when you compare it to something like a Datacap or FileNet system, then to a manual process. For example, when you have to bring in more people to do a job, it takes training, time, knowledge transfer, money, etc.
The difference when you have a system that's already set in place with automated workflow structure, you can just bring people on, show them how to use this system, then they can work the system for 10, 20, or 30 years. What I have heard from IBM is, "If a job is replaced with automation technology, you increase newer jobs to be able to run automated systems."
So, you save a lot of time and money, but the benefit is you have people who are able to run the systems, check to see if there are any errors at all, and there are a lot less errors than a human system.
I do a lot of the business analysis or project management in our company. I have seen the difference that it makes to our customers to implement a technology like Datacap or FileNet.
I would definitely recommend this solution depending on the need and situation. If they need help with sales, or something along those lines, Datacap is not the first thing that I would think of. That would be more of a CRM tool. When it comes to wanting to eliminate a lot of human error, wanting to streamline a lot of internal processes, or even if it's just cutting expenses, increasing cashflow, and revenue overall for the business, then I would recommend Datacap, and additionally FileNet.
IBM Datacap is a imaging platform that our company specializes in from a developer/architecture perspective. It is a platform, so you can write any type of capture or advanced imaging application. It produces great value and results.
We are thought leaders in the capture and imaging space. What we do is we help companies figure out how to use advanced imaging techniques, processes, best practices, and other tools, such as ODM or RPA. Our goal is to generate as much highly valuable output as possible.
Going forward, IBM needs to ensure that the output is perfect (as it can make the product) while staying true to platform's core.
The product is like every other enterprise product. Vendors (like Microsoft) should not be able to put software out there with a lot of bugs and fix them with enhanced packs and service packs. However, that's how it is.
No product is perfect. Our job, as solution architects and implementers, is to help customers get around these hurdles until they are fixed. We do it very well.
It is highly scalable. We have customers who are pumping through thousands of documents a day. If you're looking for enterprise, this is the solution because it is a platform. I can develop an accounts payable solution today on top of IBM Datacap, and tomorrow, I could be developing an HR application, which is huge.
One of our value adds is we have a very great relationship with the IBM support team. In a lot of cases, we will help IBM work the support system on behalf of the customer. From our perspective, support is great because we have the impact to help influence it, which is valuable.
The great thing about the cloud offering and tools that IBM Datacap implements are customers don't have to rely on anything. They are supported, maintained, and fix by IBM, so customers just pay and use.
Pricing needs to stay competitive.
At the end of the day, our customers have a great night's sleep knowing that their software is being installed, maintained, and developed properly.
It's an accounts payable automation. It automates the capture of invoices, extracts relevant data, creates an export, and feeding it that way.
It's resiliency. There are multiple ways of identifying what you are looking for. There are multiple export formats.
It's a platform, not a configured application, so you can do what you want with it.
They have to stop focusing on new development and stabilize the latest release. It is not stable.
We're going to a web interface, which is very common these days. However, the web interface in the latest release is not stable. This is causing issues. We are about to lose a very large client because IBM can't fix the product.
It's scalable. It can run in the background (zero users) to our larger clients (8,000 users).
The technical support is horrible. They have downsized the support teams too much. They've outsourced some of them along with some of the development, and they're just stretched too thin.
Unfortunately, this is not uncommon.
If someone is looking just to install and remedial capture, the initial setup is maybe a month.
Most of our engagements are a four to six month time frame. Not overly long, when compared to SAP or Oracle, which have about two to four year engagements. We are about a six month average. Primarily ours is more about understanding the business requirements. We'll not alter the business to fit the software. We'll make the software fit the business. It's understanding it to a finite degree.
That's the complicated part. Almost 20% of any of our engagement is talking to the client and end users, and understanding what they need.
It really only has two serious competitors in the market.
Kofax: They bundle their products, but they also license via page count, whereas Datacap doesn't. Therefore, initially Datacap looks expensive until you do a three to five year ROI. Unfortunately, Kofax has been sold maybe four times in the past three years.
The other one is Captiva, which was part of the Dell EMC portfolio, which was bought by OpenText. Having worked at OpenText, we terminated it, knowing, "That's where software goes to die." In our thoughts, the product is in a death spiral.
That is the difference between the products.
They've got to stop chasing their tails and putting new things in it. At the moment, they need to stabilize it. Once that is done, this will allow the sales reps to sell it, then they can go ahead and start doing the fancy stuff.
With Datacap, use cases are really across the board. Accounts Payable seems to be a really popular use case in terms of invoices, but we do everything from insurance to financial services, to state and local government. The great thing about capture is that every company has documents they need to capture in some format, so use cases are pretty versatile.
For the customers we work with as a partner, a lot of times the solution is geared at introducing efficiencies for the mailroom. If you think about an insurance company, where they're getting a lot of paper coming in, they traditionally have to scan it and have someone sit there and key from that image.
Datacap can take a document, maybe not having seen it before, and can classify it automatically. It can do OCR, extract all the data from it automatically. With it, they're really introducing real cost savings because the documents go straight through, don't have to be touched, handled as much, can go to the back-end systems more quickly. There is real efficiency that gets introduced through technologies like Datacap.
The user experience. One thing that IBM did a few years ago was they standardized all of their ECM products on Content Navigator, including Datacap. If you're an IBM ECM customer you have FileNet, you have Datacap, you have StoredIQ and you have a consistent user experience, user interface. It's all web-based. They moved to the cloud. They're doing things like containers. They've got a roadmap that's pretty "emerging" in terms of where they're going, that gives you that complete solution.
Reporting and analytics seem to often be something of an afterthought. With Datacap, they've started building out some dashboards, but one thing we hear from our clients a lot is, "Well, gee, we really love reports. What Datacap has is not really helpful. We'd like something better. We'd like more dashboards." That's one area where we've seen some feedback that the product could do better.
It's a stable product. We've seen other capture products on the market and our view of Datacap is that it's best-of-breed. It's a complete solution. It has its areas where it could improve, but it's reliable, it's scalable.
One of our largest clients, a large insurance company that you would know, they're ingesting 500,000 pages per day through Datacap, a really high volume. They've scaled it out to over a hundred servers, but that's the way it works. That's the design of the product: it can scale out as much as you can add servers to it.
The setup is getting better. IBM has introduced some more configuration wizards and the like. But what we always advise customers is, if you're just starting in with Datacap, bring in a partner like us to do the first one, because we've been working with this type of technology for many years. It's becoming easier to set up but you really want to work with someone who knows it, knows how to optimize it, knows what the best practices are to really make sure that you're getting the most ROI from the investment.
I think it is top of the market but there's always room to improve on it.
We use it for digitization and automation, meaning auto-classification of documents, boundary detection, and the flexibility of building a rules-based approach to building capture solutions. More importantly, we use it for extracting content from that data.
As a partner, 60 to 70 percent of the organizations we work with are in the financial industry: big banks and mortgage lenders. We find that we bring them a solution for separation of what they call "blob documents," large PDFs that contain anywhere between 100 and 200 document types. They want a solution to crawl through those documents, to break them apart, to understand the boundaries of the documents and identify them and, after that, to extract information from those documents.
The big thing these days is really the Insight Edition component and being able to build annotators to extract from literally unstructured content: paragraphs and information where there's no start anchor point to define where that data is located. There could be a number of entities in that which you have to draw information from. Being able to extract from them is really the differentiator today between this product and many of the other products like Kofax (KTM). ABBYY is one that provides something very similar, but it's not really known in the space yet.
I've always had my own gripes. Speed of OCR is one issue. It's a challenge because we have customers that have millions and millions of pages that they want this solution to crank through. In order to do that you have to have a large infrastructure in place, and that directly impacts licensing, based on the core count.
The user interfaces for exception processing can be tweaked. I commonly find that we try to tweak and customize some of those components to more of what the industry standard is. The product is still trying to play catch-up a little bit in those areas.
The stability all depends. I find that sometimes it's very stable, release after release, but I also find that sometimes you'll find bugs introduced, release after release. That's just something you just have to deal with, open a PMR to have them correct.
From a licensing standpoint, it's tailored for enterprises. It can be used for departmental purposes, but it might be overkill. It all depends on the use case. But it is enterprise, it is highly scalable with enterprise role-runners and by adding more hardware to the solution to generate more throughput of documents.
Tech support is knowledgeable. They've been quicker to respond than usual to my PMRs.
The installation and configuration are fairly simple. It's a standard install process, not overly complex. An argument across customers is that it is difficult to build solutions using IBM Datacap. I think they're continually trying to work on that.
We also provide solution accelerators that we build on top of Datacap, to accelerate those processes so it's not as complex. That's one of the reasons why we hope that people come to our company, Imagine Solutions.
I rate Datacap at eight out of ten. I'm very versed in the product, I know it very well. It's very flexible. I like the flexibility because I have worked with other products that have more of a black-box approach. Datacap is not a black-box approach. You know what's occurring through that solution inside and out, and you can tweak it or tune it.
I rate it at eight, and not a nine or ten, because we constantly have the complexity, customers telling us this solution is complex, overall. We tend to train customers after implementations of a week to two weeks of our own guided training, but it really takes someone, hands-and-feet, working with this solution for quite some time to really grasp all the concepts. We implement complex solutions and, sometimes, it's not as easy for a customer to acquire that knowledge after our departure.