Wednesday, September 29, 2010

The COLD War Rages On

The COLD war has consumed my entire professional life.  No, not as a secret agent working for "the company" and not the Cold War as in "Mr. Gorbachev, Tear Down This Wall!"  I am talking about the struggle to manage and present huge volumes of computer generated content.  The term Computer Output to Laser Disc died sometime between Sputnik and unilateralism but COLD lives on as a term for this struggle, and if you are wondering, COLD warriors never die.

This post lays some groundwork on the basics of COLD because there are fascinating topics to be explored just over the horizon.  Output management is a very significant focus within ECM due to some peculiar traits about this content type.  Did you know that high volume transactional output (HVTO) is by far the most voluminous content type managed by corporations?  The specific content we are talking about here includes statements, letters, notices, invoices, and reports.  Reports, by the way, represent a whole other conflict.

Most organizations either implement separate COLD systems or attempt to store and retrieve output from their existing ECM's.  There is a hot list of do's and don’ts that will be explored here soon.  The traditional differences between COLD and regular ECM are:

  • COLD data is transactional, and therefore fixed in time, static
  • Other content can be evolving, not static, and is more likely to be linked to workflow
  • High percentage of COLD is repetitive, think of millions of customer statements: the only data on the page that change are customer name, address and unique transactional data
  • Repetitive = opportunity to condense, stack, and "store-once-use-many"
  • COLD is way high volume, usually measured in millions of pages per day/week/month
  • COLD data is often used to answer customer questions about transactions, either internally through front-line CSR's or externally through portals
  • Because of the service angle, there is a huge impact to efficiently storing COLD documents and retrieving them with scalded-dog speed

So, you can begin to see how different COLD is to, say, Accounts Payable Invoice Capture, for example.  Its uniqueness represents a launch pad for other opportunities to consolidate systems, use emerging products like SharePoint, improve customer service, reduce costs and even to look to brand new technologies that might obviate COLD as we know it.  Stay tuned for posts next week because COLD is not dead, just lurking under some new identities.

Monday, September 27, 2010

Fire Up with Ignite!

Forgive the minor pause on ECM topics and indulge me for a moment on a global networking movement event that blew our minds last week.

Ignite is a blend of networking, ideas and social awareness, and is attended by all walks of life, but seems to concentrate on us Technocrati.  The event starts out like a regular networking thing for the first 30 minutes or so, (did I mention excellent micro-brews?) but then there are several, presentations given.  These are not your father’s PowerPoints.  Instead, they are EXACTLY 5 minutes long, comprised of 20 slides that auto-advance every 15 seconds.  The topics last Wednesday night ranged from:

-         Impact of delivering 600,000 phone books per year to residents of Fort Collins
-         How Business Can Make Health Affordable
-         How to attract, retain more clients through flat-fee billing approach
-         Go Figure: A Visual History of The Female Form
-         History of Human Kind (In 5 Minutes)

The presentations were laser focused, very funny, and asked the audience to do something at the end.  For example, the phone book presenter had the phone numbers of each of the phone book publishers available for everybody to call.  By the way, does anybody actually use phone books anymore?

One of the interesting people I met included Betsy Craig, Founder, MenuTrinfo.  Her firm provides restaurants with the nutritional information of their recipes.  With New York City’s and the State of California’s recent mandates, many larger restaurants are scrambling to provide dietary information about their dishes.  Her firm employs nutritionists and dieticians that turn a menu into detailed information that can be delivered over the internet.  Is there an app for that?

Miles Kailburn of Old Town Media had some very interesting comments and ideas on SharePoint, which I will explore in a future post.

Check out what Ignite is all about: http://ignite.oreilly.com/

Monday, September 20, 2010

Mailroom Automation is for Lawyers; TransPromo is for the Birds

No sooner than the day after my post on complex document automation, did we run into yet a new wrinkle.  Some of the post processing experts talk about mailroom automation and the impact this can have on 1:1 marketing, trans-promotional marketing within documents, and the Holy Grail - highly personalized, high volume documents that sell more of your products.  Why is everybody talking about this?  Does anybody actually do this in production?

We have no idea, but the allure of being able to cross-sell large numbers of customers on related offerings through your billing process sounds like a money printer.  The question that always looms though, is whether anybody actually looks at anything on a homeowner’s insurance bill, for example, other than “AMOUNT DUE”?

Well, a very large insurance company has another take on mail automation that has a lot more to do with attorneys than marketers.  One of the benefits of really nailing down your mailroom automation is a concept known as Registration of Mail (ROM).  An interesting thing happens in insurance when you mail notices of cancellation or bills.  Sometimes, insureds, or soon-to-be non-insureds, claim that they did not receive legal notices or bills.  Think of the impact when a newly cancelled non-insured drives his car into the Dry Cleaners, literally.  How do you overcome this if you have millions of folks that you insure?  You likely do not use FedEx or Registered Mail to deliver all of these documents.  Instead, through a registration process with the US Postal Service, and through very careful sequencing of documents by zip code + 4 + last name sort, you can put in place a process that ensures delivery to customers.  Or to put it another way, you can ensure that the courts will believe that it is highly likely that the intended recipient received the critical document, on time.  This translates into an effective evidence vehicle in case litigation is ever required, and at the same time, reduces postage expenses to the insurer. 

I know, it sounds so pedestrian, doesn’t it?  Just remember though, it was income tax evasion that brought down Al Capone.  It might be that the hype of trans-promotional marketing and 1:1 personalization will finally bear the promised fruit.  In the meantime, the attorneys will find a way to use auditable technology to win, all the time.

Friday, September 17, 2010

We Do the Hard Stuff

People in my company are often challenged with how to give the best “elevator” description of what we do.  We are not an application.  We are not really middleware.  We don’t solve any single problem.  But we do work with companies on very vexing problems, the kinds of problems that seem oh-so simple at the start of the project.

The abstract view of what we do is complex document automation.  More simply put, “we do the hard stuff around document generation, archiving, printing and presentation.”
A Simple paradox in life is that what at first seems easy can sometimes turn out to be the hardest to accomplish.  This is very much true in the world of document automation.

Why is it that we can construct new secure portals for outside access to sensitive content, find ways of locking down even Blackberries, but when it comes to reliably putting pixels on paper, the wheels can literally come off?  The answer is not that it is complicated - that goes without saying.  What we are facing is the gap of where printing is in our heads vs. where it really is circa 2010.  Let me elaborate with a quick for-instance:

A large investment firm needs to consolidate customer statement print from two centers into one and create mailings from these systems into a single, mail piece for hundreds of thousands of customers.  There is a huge opportunity to drastically reduce costs by consolidating centers.  Ok, you say, just "merge" the print and maybe transform the print to the format supported by the newly consolidated center, right?

If only it were really that simple!  Over the years, the guys and gals in the print shop have implemented a slew of tricks to improve efficiencies and reduce costs, especially mailing costs.  Here are some of the tricks that need to be accommodated in order to consolidate the print into the new center:

  • Recognize the arrival of print packages, and unzip them
  • Parse related .csv file that contains instructions for manipulating the print
  • Translate the .csv information into a specialized data format for downstream mail systems
  • Consolidate multiple print files into a single print file
  • Add other metadata for downstream processing, archiving, discreet presentation
  • Transform or normalize the print into a single format supported by the printers
  • Add Barcodes to the printed pages so they can be read by envelope inserters
  • Add banner and trailing pages to ensure that batches of print are kept separate

These are the kinds of requirements that drive our business, and sometimes drive our customers a little crazy!  Let me know what you think – if you have any war stories that you can share, I’d love to celebrate or commiserate with you.  DK

Monday, September 13, 2010

Enterprise Content Capture

The ECM Trends I follow include Web Content Management (WCM), High Volume Transactional Output (HVTO) and ePresentment. The field of document imaging and capture is a little off my beat, but we’ve been seeing some significant advances in capture that make sense to talk about.



In the old days, organizations would use capture solutions to scan, index and OCR documents, and then “release” the documents and metadata to the backend archive or workflow solutions. Very simple: Capture, Index, Load. Often, organizations would supplement this process with customized management systems to ensure that the expected number of documents was actually “captured”, that the metadata contained the right number of fields or characters, etc. Additionally, as requirements around capture expanded, additional custom utilities would be added to unzip the odd file, or notify the user upon failure. These systems were usually confined to one or two departments, such as Accounts Payable for incoming invoices, or Claims for insurance claims, for example. In short, the dirty secret in the old days was that the Enterprise Content Management (ECM) was rarely “Enterprise” but more departmental.


Fast forward to today. Now, organizations are not only scanning and indexing document content, and loading ECM systems from many departments; they are truly deploying content capture across the enterprise and linking the capture process to Line of Business (LOB) applications, and it is completely bi-directional. In many cases, data is being extracted from LOB systems to assist in populating or validating the indexes for the documents. Content is also being included in LOB processes for decisioning, so there is now a requirement to use document content as part of a process. Plus, today, we have web applications that are using the documents to do their thing. So, today, it is not so simple: Capture, Index, Validate, Load, Extract, Notify… you get the picture!


So, the old custom code from yester-year no longer cuts it for truly enterprise-scope deployments with multiple departments loading and extracting content and using webservices as the glue to get it all done. What is needed is an enterprise content capture approach that has the following:

• Single Infrastructure with multiple events and processes

o      Webservices, http(s), ftp(s), email, sockets, command line, others

• Load high volumes of document content into the archive

• Validate successful loading

• “Rollback” the load in case of failure or inadvertent load

• Easily connect to multiple backend systems, databases for extraction or loading

• Ability to do mundane things like unzip files

• Normalize or transform file formats (transform to PDF, or .CSV or TIFF, for example)

• Extract content from the archive to pass onto other web-based processes

• Ability to easily notify users or administrators of errors in the process

• Manage it all from a web-based GUI with process flow design and preflight



So What? The cost to large companies to constantly replicate custom capture systems is huge. We’ve seen some organizations looking at spending several $ million to extend departmental capture to just 1 or 2 additional departments. These processes are complex, involve a lot of users and custom code, and the risk of getting it wrong is too great. And, don’t even get us talking about compliance as a reason to do things once, and do them right.



A company I’ve found to be a thought leader on ECC is Vega ECM. They have been involved for years with organizations on content capture approaches, and as consultants, helped write a number of the custom approaches I alluded to above. Vega now sees this as a discreet discipline within ECM, and is helping organizations move from the custom approach of yester-year to more modern, single-infrastructure approach bulleted above. Check out http://www.vegaecm.com/ for more.

Monday, September 6, 2010

Storage Is Not Cheap!

For years, there was sort of a dark joke going around that the cost of storage was dropping so quickly, that there was no need to think twice about ordering that additional Centera Drive, “just open up another Clip-ID and go!”

It is urban myth that storage is cheap. Maybe the 1 TB USB Drive you can buy at Best Buy costs less than a full tank of gas, but that nifty backup drive does not have to live in a redundant, arrayed, hot-site-backed- up data center with security I am not allowed to describe in any detail. Here is the math for reducing storage for one firm we’ve helped recently:








Above figures depict the cost difference to store AFP compared to PDF for insurance policy and billing documents at a major insurance company.

We encounter this all the time when working with clients that are making tough decisions about whether to store in PDF format or store in a “native” format such as AFP, metacode or PCL, and dynamically transform to PDF when the documents are retrieved.

Several important facts:

• If you are storing high volume transactional output like statements, policies, EOB’s or 1099’s, the biggest part of that document are the repetitive graphics, fonts, and overlays used to make the document attractive when printed.

• By storing the native format, you can store those repetitive elements ONCE, version them, and then re-combine with the variable data upon retrieval.

• This makes even more sense when you consider that on average; only about 3-5% of any transactional document is ever retrieved from your archive. Why redundantly store all those expensive graphics over and over?

• Retrieval performance for dynamic retrieval and transformation has improved so much that users typically do not even notice that their document is being dynamically rendered.

As we all try to find our way out of the Great Recession, ideas like Storage Reduction are getting the attention of senior managers looking to cut costs.

ECM Trends - Maiden Voyage

OK, this is very frightening! This reminds me exactly of the time when the editor of my college newspaper asked me to write sports articles, mainly short snippets about how our Alma Scots faired against their bitter rivals. And, for some unexplained reason, I said yes.


Maxine Buttons, editor-in-chief of the Scotsman was wonderful, persuasive, and had an intellect that could instantly summon prose on any topic, all with the Who, What, Where, When, Why included in the first paragraph. I, on the other hand, labored over how to describe anything that happened in the Alma soccer team loss to Albion, 2-1.

So, as I set out to describe what is happening at the project-level with Enterprise Content Management (ECM), it is with this humbling experience that this blog commences. And yet, there are some insights about this field that are not being talked about, ideas that are not being developed, and secrets that, if shared, could benefit us all.

ECM includes a whole host of technologies and methodologies for organizing, storing, categorizing, discovering and retrieving vital corporate or organizational documents, website pages, videos, in short anything created by or for the organization. My experience, and therefore the focus of this blog is primarily on computer generated content, or High Volume Transactional Output, but because I have over 20 years of experience in this field, I often hear about eDiscovery trends from my clients, or am privy to specifics on developments in the document capture field, for example. The ground rules of this space will be no names will be used at any time, other than the talented Maxine, above, but that genuine insights, with specifics, are fair game.

Let me know what you think at any time – your feedback and input are like nuggets of gold, and your payback will be the collective intelligence of us all!