Modified by Scott Blau email@example.com
I have been involved in document capture for 26 years - since well before it was even called "capture." I am often asked about how I came to found Datacap. So taking advantage of my impending exit from the stage (read on for more!), I thought I would share some background on the founding of Datacap as the first true document capture software company.
The microprocessor revolution was in full swing in the mid-1980s when I joined a hospital information systems startup. We concentrated on data collection in operating rooms using what was then cutting edge: IBM PCs networked directly together with TCP/IP. We got sterilized computers into the operating rooms... but we struggled to get nurses to use them.
Turns out that not only was the nursing staff more comfortable working with pen and paper (usually on a clipboard), but they resented having to literally turn their backs on the patient - particularly when they were being asked to enter what was essentially inventory data for billing purposes (the materials and instruments being used during the course of the operation). We also had a back up system in case the PCs or the network didn't work (which was often). It used... paper.
It was in this context that I was struck down with a very nasty virus: chicken pox. You may think of chicken pox as a childhood disease with some discomfort - or even as a great opportunity to play hooky from school - but adult chicken pox, as I quickly learned, is an entirely different beast. I ran a high fever, and for several days I literally could not raise my head off the pillow.
At times I became delirious. And it was in a delirium that Datacap was born. I had visions of paper forms filled in by nurses in the OR dancing around... but, more interestingly, of the data on the pages coming unstuck from the paper and floating off. I saw individual characters very clearly and how they could be segmented to be understood by the computer as data.
Once the fever had burned itself out, I got back to work. One day not long after, a guy I had brought in to help us with some of our tougher user interface challenges showed off an early document scanner. He was building a driver to run the scanner from a Mac for another client of his. It seemed an amazing piece of equipment... and it struck me right there that if we could scan the nurses' sheets, then we could segment the characters and turn paper into data!
Once the wheels started spinning, there was no turning back. Along with Noel Kropf, the guy with the scanner, we founded Datacap and we set to work building our first product, Paper Keyboard, releasing it in 1989. It all seems so inevitable now, but at the time there were nothing but hurdles to overcome: porting to Windows 3.0, adding machine-print OCR, tying multiple machines together to distribute the work, etc., etc.. It kept a growing team of developers busy for the next two and a half decades.
And, of course, we faced increasing competition as the "forms processing" business became a recognized speciality in the quickly gowing document imaging industry. Some vendors edged into scanning and data entry automation from related areas like manual data entry (Textware, later Captiva), while some started the long transition from hardware to software (Kofax).
Eventually, Datacap became part of IBM in 2010, giving us the opportunity to put down a global footprint. What started off literally as a “vision” in a fever, has become a global reality, used by customers worldwide to ingest millions of pages each day. In some ways, for me, that fever never passed. It has energized me for years – and I like to think I have “infected” a few others. If so, then maybe my job is done. I can let a new generation of visionaries take what we have done with IBM Datacap to a new level. That's part of the reason why I decided to step aside from my current role into retirement at the end of this month, not long after I push the "Publish" button on this blog.
I still have a vision for the future of document capture, one that is increasingly mobile and distributed, and one that will make the steady transition from "on-premise" to SaaS for many customers. But I'm sharing it with you now so that you can help make it a reality while I spend some more time on my bike and doing the many things that I haven't had the opportunity to do since I first caught the Datacap bug!
Whatever I’ve done, I’d like to thank the hundreds, and probably thousands of individuals that have made the document capture space a thriving business arena. Whether you were with Datacap, with IBM, with one of our many partners, or even with a competitor (ha, I know you are reading this!), without all of you, we would not have such a vibrant and successful capture community!
Scott - @CaptureGuru
Modified by Scott Blau firstname.lastname@example.org
As a continuation from my previous post, here are some fundamental questions to ask yourself - and others - as you embark on a distributed capture endeavor:
Is it “Usable?”
An intuitive user-interface is essential to facilitate distributed capture. Typically, the people receiving documents are customer-facing, not dedicated and trained capture operators. The solution should provide a clear and simple series of steps to that assure a legible document image…
Can it be “Read?”
A poor image quality or, worse, partially-captured document, will quickly undermine the benefits of distributed capture, especially downstream when it comes time to extract data with optical character recognition (OCR). This is where most mobile telephone cameras struggle to create high enough quality images to avoid laborious manual effort later in the process. For a step up in quality, select a portable scanner – some are no larger than a thick ruler – that attaches to a laptop or mobile device.
What document is it?
The first, most important, piece of information about any scan, is the identity of the document itself. Is it an application, a claim, a change-of-address, etc? That question might be answered by manual input from the person who scanned or took the picture of the document, but it also might be automated through automatic document classification. Remember, your mobile and distributed workforce are not trained capture professionals, so take a belt and suspenders strategy on this one…
Is it Accurate?
Determining the accuracy of content extracted from a document is of prime importance. Whether the extraction is manual, or automated with OCR, you need a set of checks and balances to assure users that the solution can be relied upon. For example, if the software is uncertain, how does it notify a user, and which user is it that gets notified?
Is it Safe?
The security of data is essential to consider, especially when handling customer or other sensitive data. Distributed capture must be considered moving capture into high-risk environments. Make sure you understand what the risk exposure is if a mobile device is lost or stolen in the field.
Is it Faster?
The speed at which the captured document is transferred from the mobile device to your repository or LOB system determines the speed at which it can be processed by the application. The old saying, “a chain is only as strong as its weakest link,” comes into play here. If there is, in fact, a bandwidth limitation for remote users, then the advantages of capturing remotely may be lost in the transfer.
Is it Capable of Handling Anything a User Throws at it?
There are always exceptions and how you manage them is the test of a capture system. Can you add attachments? Can you add a new document you weren’t expecting? Can you annotate a document or route it to a supervisor for review? The closer you are to the customer, the more exceptions you will encounter, so make sure you have the flexibility to handle the unexpected.
Will it work for me?
In most cases, a mobile capture solution will both archive the document images, and route them into a line of business system – as fast as possible for customer satisfaction. For example, an invoice, resume, or contract will be sent to the ERP system. An insurance claim will be forwarded for adjudication. A loan application may link to a case management system, where underwriters will review. A medical document will be appended to the patient’s electronic health record. Make sure your distributed capture system can connect to your business systems and deliver image and data seamlessly.
After all these years in the capture business, I thought things had settled down. People have been saying that document capture is a “mature” technology. And, of course, it is, but the world is changing around us, creating new opportunities. So don’t be shy: if you see a way to shorten the cycles, to deliver better customer service, to improve vendor relations, or to change just about any existing process by capturing documents sooner at distributed/remote locations, then take advantage of the opportunity. Just ask the right questions - and get credible answers – as you navigate to a successful implementation.
Note: An earlier version of this post appeared in April 2013 on John Mancini's Digital Landfill blog.
Follow me on Twitter @CaptureGuru
Modified by Scott Blau email@example.com
It’s a given: the sooner you convert a paper document into an electronic image, the faster, more accurately, and less expensively you process it. Obvious though it may have been, over the 20+ years I’ve been in this business it’s not been an easy insight to act upon.
In the era of MFPs (multifunction peripherals), mobile phones and mobile data plans, it’s easy to forget how tentative data connectivity was even a short time ago. Even in a commercial setting, banks with branches, insurers with independent brokers, in fact, any organization with far-flung activities, all had big concerns about wide-area bandwidth. Scanning of documents and sending them “over the wire” from remote locations was seen as a luxury.
That perspective is changing – fast.
Converting a paper document to digital image as soon as the document is received, or even created, is a strategy now within reach of most organizations, in most parts of the world. It's called distributed document capture. It’s different from the old model of centralized capture, where everything is sent to a central processing center.
The good news is that there are now low-cost desktop scanners, mobile scanners, multi-function peripherals (MFPs), and more than a billion smart phones worldwide that can operate as a capture device. The bad news is that it’s not so simple as simply snapping a photo to be successful with distributed capture. Before you invest in a solution, you need to prepare yourself by asking some key questions... I'm putting some together to share with you in my next post.
Follow me on Twitter @CaptureGuru
Modified by Scott Blau firstname.lastname@example.org
It's time to start planning your agenda for Information On Demand 2013 - aka "IOD," in Las Vegas. Whether you are in IT, Operations, or Finance, IOD is a great networking opportunity: meet with peers, industry experts, and influence the architects of your current solutions. Choose between business, technical, and leadership training sessions, or use the event to expand your understanding of Business Analytics, other Enterprise Content Management (ECM) technologies, and Information Management. There are also special events with todays thought leaders. You will be encouraged to “Think Big,” but maybe just as important, you can also learn how to “Think Fast.”
The main reason to go to IOD? Capture, of course! We're putting the focus on capture in the context of "real-time imaging." What's real-time? That's the time you - and your customers - expect things to happen when they have a smartphone in their hands. Mobile is coming to capture very quickly now. Don't believe me, then come to IOD to see for yourself. We'll be showing that and related distributed/branch capture capabilities and solutions. You'll see what is available today... and if you pay close attention, we'll give you a sneak peak at the future!
Here are some specific real-time imaging sessions to pencil into your agenda... and there are more to come!
EIC-3440A: Time is Money: Coca-Cola Realizes Process Improvements with IBM Datacap, Speaker: Thomas Fantroy, Coca Cola Refreshments, Manager Imaging & Workflow Solutions\Monday, Nov. 4, 10:15 – 11:15 AM, Lagoon U
EIC-1815A: Mobile and Multifunction Peripheral Transactional Capture to IBM Datacap and Enterprise Content Management, Speakers: Anthony Vigliotti, Notable Solutions, Date: 6th November, 2013, 4:30 PM-5:45 PM| Location: Lagoon IJ
EIC-1667A: What's New with Mobile Capture, Speaker: Mattias Marder, IBM, IBM Research - Image Processing and Computer Vision, Thursday, Nov. 7, 10 -11 AM, Lagoon GH
ECG-2224B: Content Integration: A Success Story (Mobile mortgage capture at National Bank of Canada) Speaker: Alain Foisy, National Bank of Canada, ECCM Practice Leader, Tuesday, Nov. 5, 3:00 – 4:00 PM, Lagoon F
Also, be sure to take advantage of the once-a-year opportunity to meet 1:1 with IBM executives, subject matter experts and innovative IBM Business Partners. I'll be there, but you an also talk strategy with other ECM imaging business leaders, such as Brent Bussell, Feri Clayton,Brian Phelps, and Rick Gawronski. Or take a deep dive in to Document Imaging and Capture with experts from our product and technical teams, including Tom Stuart, Ben Antin, Jim Reimer, Charles Wiecha, Bud Paton and Noel Kropf.
Learn more about Information On Demand
For ongoing IOD updates, follow me on Twitter @CaptureGuru.
Modified by Scott Blau email@example.com
I'm just back from a trip to India. Until fairly recently, I would never have imagined significant opportunity for document capture in the very place where outsourcing of data entry has been most successful. That's relevant to the document capture business because when a document is "captured" two things happen:
- the paper document is digitized (usually scanned, but sometimes an already electronic document is converted to a standard format), and
- data is extracted from the document - either manually or using OCR - so the document can be filed, and sometimes to populate a line-of-business application.
Call it what you will - indexing, verification, keying - it is a data entry requirement that allows a document that has been digitized to be sent to locations around the world where labor is less expensive. It is exactly in this type of work that India has excelled. A vast, trained workforce has taken on the tedious task of manually extracting information from documents. Even compared to "automated," OCR-assisted data entry that requires relatively expensive labor in North America or Europe, a very competitive alternative has been to take advantage of the much lower-cost labor pool in India (and other countries) to manually enter data, without the help of automation at all.
So why was I in India? To some extent, you can say that the success and breadth of outsourcing initiatives over the last 20 years have changed the underlying economics. Although labor continues to significantly less expensive in markets such as India, China, Philippines - the usual suspects - costs have gone up substantially. They have gone up enough that it is no longer a given that throwing more manpower at a problem, such as manual data entry, is going to be less expensive than investing in automation technologies to help assist in the effort. Many organizations are coming to the same conclusion.
Even banks with far-flung operations and massive workforces are exploring ways to automate aspects of the document capture process: the volumes of documents to be captured are staggering once a bank wades into the world of branch capture. (My thoughts on how branch capture is technically something new in document capture: http://ibm.co/13i74bl.) Automation not only reduces costs, but speeds up the process, ultimately helping improve customer services… and most importantly, customer satisfaction. (And if you are skeptical that customer satisfaction is the underlying benefit of document capture, let me try to convince you: http://ibm.co/10bwmsJ.)
Put another way, in large-scale document capture operations, there is a premium on reducing complexity, including the number of people involved. Globally, the Holy Grail is to grow the number of documents being captured, while meeting that growing need with existing staff.
From my perspective, document capture has come of age when it is being adopted globally, even in markets traditionally noted for the low cost of labor.
To continue the conversation, connect with me on Twitter @captureguru.
Modified by Scott Blau firstname.lastname@example.org
4 Non-Trivial Questions to Ask before Committing to Production Document Capture
In late 2009 and I got a call from the brother of a good friend. He was a researcher at IBM's Watson Labs - soon to became famous for the "Watson" artificial intelligence engine that spectacularly beat the top humans on the trivia game-show, Jeopardy!
My friend was trying to solve a problem and thought that my company, Datacap (the acquisition of Datacap by IBM was not even on the horizon at this point), could help, since we specialized in optical character recognition (OCR) and related document capture technologies.
I said, "great, let me ask you 3 or 4 questions about what you are trying to do:
1) What is the volume of documents/pages/images you need to process per day, week, month, or year?
2) What data do you need to extract from those pages, any special considerations to take into account?
3) Are the pages consistent in format, variable, something in between?"
He said he had 5000 pages. Clearly to him that was a big number, but he was a bit deflated when I asked, "is that per day?" In the production document capture business, it is definitely common that a volume like that may be literally processed "before breakfast."
But 5000 pages were all he had. Not every day or week, or even every month, just once. I was a little skeptical, but I wanted to learn more.
He needed to extract information from an English language pronunciation guide. He wanted to read the word to be pronounced, and then the linguistically precise definition of the pronunciation, including diacritical marks (accents) commonly used in those definitions. In other words, this was not just straight English language OCR. My skepticism increased.
I wasn't surprised when I next learned that the pages were not at all consistent, that the definitions for a specific word could wrap from one page to the next, or that the pages to be scanned were in bound books...
That was it. Did he really expect to use a production capture product to process - one time - 5000 pages with specialized text and words on them and no fixed format? Well, yes, he did. He had a real challenge and his expectation was not unreasonable... it just is not what production document capture is about.
Those three questions can help anyone quickly assess a document capture problem. In this case, the answer was simple, but perhaps wrong. I advised him that it would not be economically feasible for him to invest in production document capture, but in giving that answer I missed a great opportunity.
Turns out I should have asked a 4th question, "why do you need to read a pronunciation guide?"
I learned later that my friend was working on a major artificial intelligence project, one that would need a computer capable of blurting out words under extreme time pressure. He was, in fact, working on giving "Watson" a voice. It was that voice, having been trained to enunciate thousands of words, that went on prime time to beat the best human players at a live game of Jeopardy!
He eventually used a desktop OCR program and a lot of patience to translate the pronunciation guide from paper to something Watson could understand. Although my 3 questions helped me quickly assess the value of the opportunity, by skipping the 4th question, I missed the opportunity to brag how Datacap helped to give Watson a voice!
Is production document capture and imaging right for you? Click here to learn more on using capture solutions.
Modified by Deepthi Nagarajan DEEPTNAG@IN.IBM.COM
Blog Post by Scott Blau,WW Director of Document Capture,IBM ECM
Historians and anthropologists debate the topic of “when did intelligent life on earth start?” But I sometimes wonder just how intelligent we are when I see the nutty things that humans do. I certainly scratch my head at the way we sometimes make our lives so hard for ourselves in the way we structure our activities.
Is it really intelligent to pave over so much of the earth? Are we being smart not putting more effort into alternative energy sources? And on a much more immediate level, is it really so intelligent of us to continue to be swapping paper documents and filing them in filing cabinets, or even printing an email… so it can be sent down the hall to be scanned?
It is probably a “no” to each of those questions.
I’m not an expert on ecology or sustainability, but I’ve been around the document capture space for long enough to know that there still is a lot of room for us to get smarter. One of the ways of doing that is to stop thinking about document capture in isolation, separately from the storage and retrieval of images after they’ve been captured.
When I first started out in document capture in the late 1980s, it made some sense to think about scanning a document, extracting information from it and then forgetting about the image altogether. But with advances in network bandwidth, storage, and database technologies, it’s rare these days that anyone scans without saving images. And that’s smart.
But to squirrel away those images where they may not be easily found, are not properly indexed, or are unconnected to business processes, isn’t particularly clever either. And certainly processes which cause electronic documents to slip back to paper (through printing) only to be scanned again is on the dumb side of the scale.
Intelligent Imaging is about bringing to bear all the capabilities that have developed over the last 20 or 30 years to take documents, whether they are originally electronic or paper, extract information from them, make them available, and, finally, tie them into useful business processes.
Here’s a man-in-front-of-a-whiteboard video where I explain more about intelligent imaging – all in under seven minutes.
Blog post by Scott Blau, WW Director of Document Capture, ECM, IBM
Admittedly, capturing invoices
automatically is not as exciting as riding Disneyworld’s Space Mountain roller
coaster ride, but it’s not as scary either. Why? Because you have more control. When you ride Space Mountain, you are
strapped in like a piece of luggage and you spend half the time in the
dark. IBM Datacap Taskmaster Accounts Payable
puts the operator in the driver’s seat with lots of tools – available at a
single click – and you’re never in the dark.
Leveraging IBM’s Intelligent
imaging approach, Datacap enables operators to view scans, faxes, emails and
email attachments all in the same format, so it doesn’t matter how your
department receives invoices. As you can
see in this quick demo, Taskmaster extracts data automatically, applies the
vendor ID, and enables matching against purchase order line items, so you can
help facilitate a three-way match in SAP or Oracle. And watch how easy it is to receive a new
invoice and set it up – without building templates or programming.
So here’s a proposal: automate
invoice processing with Datacap to cut your invoice processing time and cost in
half, and then, with the extra time you have back, you can make plans to fly to
Orlando and wait in line for Space Mountain.
View the demo on the IBM
ECM You Tube Channel and for more information click
by David Jenness, Market Segment Manager, Document Imaging and Capture, IBM
After 15 years of looking after Datacap’s marketing
interests, I was pretty sure that I could no longer be surprised. When it comes
to document capture, it’s easy to feel like you’ve seen it all. Yet, IOD 2012 was filled with surprises – and
the good the kind. Since joining IBM’s
ECM division in 2010, the Datacap product – with its open, rules-driven
architecture – has inspired many, it seems, to extend its capabilities in all
kinds of directions. It was at IOD,
where the extent of all that extensibility became apparent to me.
I had the pleasure of hosting a luncheon at IOD 2012 called
“The Seven Secrets of Success for Capture Engagement.” The idea being that we would invite Datacap
users and prospects to a delicious lunch and then show off seven new things you
can do in the Datacap environment that help our customers delight their
customers. I started asking around about
new Datacap applications and, suddenly, I was wondering how to keep the list to
Within the IBM technical community, our developers had come
up with cool things like integrating Datacap into a Fujitsu network scanner so
that anyone could walk up, push a few buttons and start capturing documents
using the scanner’s touchscreen – even for verification of data. We also demonstrated
how to add new documents into the Datacap Entry-Level product with just a few
clicks – setting up sophisticated document ID and data validations in the
process. And Datacap co-founder and
software architect Noel Kropf acted out the part of a delivery driver –
complete with brown shirt and shorts - who saves his company time to bill by using
Datacap with an iPhone to capture signed delivery documents on the spot.
If demonstrating Mobile Capture was exciting – and it
certainly had our audience of 60 customers, business partners and IBMers
sitting up and taking notice – what got me revved up was seeing what our
business partners had done.
IBM partner Databank showed
an application they designed and installed at a bank to accelerate loan
approvals for customers, which integrated a Fujitsu network scanner, Parascript
advanced handprint recognition, and a real time workflow to enable a regional
manager to support a branch request in minutes.
EDAC Systems, which has
developed several applications with Datacap, demonstrated some enhancements they’ve
made with image processing to improve text recognition – even for handprint –
to enable correspondence tracking, among other uses.
European partner xft showed off its certified connector to SAP, which allows Datacap to “talk” to
SAP in real time – for PO line item reconciliation during the capture process
and facilitates a smooth handoff to SAP of captured data.
Imagine Solutions, IBM’s 2012 Excellence Award winner, showed a live demo of their solution for mortgage processing with Datacap that
leverages IBM Content Classification to correctly identify and classify mortgage
documents in a batch, which is as close to a David Copperfield magic trick as
we had in Las Vegas.
All in all, we showed off 8 solutions at our luncheon and
could have shown another 8 if we had the time. Tritek Solutions has built a
human resources capture solution, Miria Systems has a proven P2P solution, CM Mitchell Consulting has developed
capture solutions for Oracle, and CGI has built a complete
healthcare document management solution with Datacap and Production Imaging
Edition, called Sovera. Furthermore, Magic Lamp Software, Neocol, and R2K have enhancements of their own that would have fit nicely in our lineup of
solutions for “capture engagement.”
For me, the “Seven Secrets of Success” turned out to deliver
unexpected surprises about the creative and technical prowess of the new
“ecosystem” of partners who have taken Datacap in new and interesting
directions. I certainly did not list all the enhancements available in this
short report and I can only imagine what new solutions are being put together
But as they say, “surprise me.”
Guest Blog by David Jenness, Market Segment Manager, IBM Document Imaging & Capture
Since IBM acquired Datacap two years ago, there has been an enormous effort to globalize Datacap Taskmaster Capture, the intelligent document input solution, expanding its reach from English language capture to the world’s languages. With every new release, Taskmaster adds new languages, and today it supports nearly 30 languages worldwide.
Released in August, Datacap Taskmaster Capture v8.1 has added Cyrillic and Simplified Chinese to its list of supported languages. What makes these additions significant is that neither of these languages uses the standard 26 letter Latin alphabet.
Cyrillic script can be traced back to the Greek uncial script (with additions from the Glagolitic alphabet, in case you were curious). Cyrillic is one of the most-used writing systems in the world. It is the basis of alphabets used in all of Russia, as well as Serbia and Bulgaria. As of 2011, nearly 252 million people in Europe and Asia use it as the official alphabet for their national languages. About half of those people are in Russia.
Simplified Chinese characters are standardized characters that the government of the People’s Republic of China has promoted for use in printing since the 1950s. Today, these characters are officially used in mainland China and Singapore. Simplified character forms were created by decreasing the number of strokes and standardizing the forms of a sizable proportion of traditional Chinese characters. Is it an important language? Let’s see, only about, oh, 1,349,313,700 people are using simplified Chinese in print today.
Between the Latin alphabet, Cyrillic and simplified Chinese, Datacap Taskmaster Capture can recognize the alphabets used by 51% of the world’s population spread across every continent. Even at this writing, new languages are being added. But this process of globalization is fascinating, because not only does Taskmaster need to accommodate the wide variety of symbols that represent the world’s many languages, but around the world, people also use different formats for such things listing dates, telephone numbers, and currencies. This means that all the internal rules for formatting and for automated validations that Taskmaster performs to test the accuracy of data also need to be updated and adapted to each language.
For more information about the IBM Datacap Taskmaster Capture v8.1 Release go to the IBM Datacap Taskmaster Capture webpage
. Visit us at the Information OnDemand Conference 2012
for exciting sessions and product discussions around IBM Capture and Content Management solutions.