Examination of Variable Data Printing (Part 1)

By | July 14, 2009

Editor’s Note Nicholas Barzelay is a recent graduate of RIT School of Print Media’s graduate program. Mr. Barzelay’s research area was variable data printing. He is the co-author of Upstream Database and Digital Asset Management in Variable Data Printing (Executive Summary available here). Mr. Barzelay is working on two books on the subjects of data management and Variable Data Printing. He is sharing some of his work to get industry feedback.

Insights on VDP

By Nicholas Barzelay

Early VDP

What we now call “transactional” variable data printing has its origins in the earliest accounting and financial programs: Accounts Receivable, Invoicing and Billing, Accounts Payable, Demand Deposit Accounting, Financial Asset Reporting, and Explanation of Benefits for insurance.

Initially these kinds of program applications were run on mainframe computers.  Later they were modified to run on mini computers.  Eventually they were migrated to multi-tier enterprise scale client-server applications.

Printing

Early VDP was generally output on chain impact printers.  The addressing and transactional data was printed on continuous-feed preprinted forms.  Identification cards were printed in a similar manner on heavier card stock.  Background designs for print stock were preprinted on an offset press.

The development of digital printers ultimately replaced impact printing.  However, it is interesting to note from a programmatic standpoint, the migration from impact to digital printing was accomplished by changing out printing hardware with little affect on application program code.  It was not until the advent of desktop publishing that VDP began to evolve into a more design intensive graphic product.

Early Infrastructure

Infrastructure associated with early VDP (transactional printing) was complex, but did not present the integration problems of current system infrastructure.  Processing was centralized on a mainframe computer running a single standard operating system and integrated application packages certified for the operating system.  Standard programming languages and data management applications provided limited choices, but also simplified the number of required skills to a few.

CRT’s (cathode ray tubes) connected directly to the mainframe and provided online viewing of data.  Printers, also directly connected to the mainframe, provided printed output.  Integration issues were localized to the mainframe and managed by system programmers who did their programming at the operating system level.

Infrastructure & Applications Now

Client-server system environments have decentralized processing in a number of ways that complicate system integration.  In the distributed processing environment of client-server, different physical processing units may be running different operating systems and application systems that may not readily talk to each other.  Units are loosely coupled via network connections.

This means that more expertise is required to keep the entire system running than a single set of tools and skills.  This may easily translate into additional staff requirements – knowledge to manage the network, knowledge to support varying hardware, and application knowledge.  Application knowledge is similarly diverse and specialized: database management, digital asset management, and graphic design software, to name just a few.

An installation’s applications need integration to share each other’s data and content.  Those applications have become more complex than the original transactional applications related to accounting.  Beyond the inclusion of digital imagery, many VDP solutions require matching and merging disparate data from different sources, drawing conclusion and developing business logic to meet solution objectives.  Solution objectives extend well beyond accounting requirements.

Those objectives include maintaining customer relationships, marketing to new customers, the application of pricing and discount models, the movement of aging inventory, and a host of other information communications.  These must be accomplished in a highly competitive environment that ultimately demands the measurement of cost-effectiveness in order to obtain optimal results.  The most successful VDP solutions are closed loop.

Conclusion

Therefore, simplification is a necessity.  Simplifying does not refer to VDP solution results (an all too often approach).  Rather, simplification means removing complexity from the workflow and integration path, beginning with data and content storage and progressing through data stream generation, document creation, and production.

The best way to simplify is to identify basic processes and tools that are understandable and easy to use.  This may mean temporarily abandoning applications with all the “bells and whistles” in favor of a few small components that are easy to understand and easy to connect together (integrate).

A variety of very sophisticated VDP solutions can be created using a few fundamental tools.  This may ultimately mean that a company can accomplish a greater variety of increasingly complex VDP solutions than those previously generated with the more expensive tool sets that may at length be relegated to producing “fancy” mail merge.

A related benefit of simplifying the workflow and the VDP development system is in the area of required skill sets.  It may be easier and less expensive to find people who can use the basic tools and then grow their expertise, rather than recruiting people with fully matured skill sets that can facilitate the most complex integration workflows walking in the door.   People with matured skills are likely to be few in number at premium prices.

Share this post

 

9 thoughts on “Examination of Variable Data Printing (Part 1)

  1. Michael J

    Thanks for a much needed evidence based summary of the development of database publishing to VDP. Hopefully it will help give some context to the widely used acronyms that are more about marketing than about understanding.

    I look forward to Part 2. The points you make are just the right ones. To me it points to the next stage which I think will be Cloud connected output devices. We’ve already seen the movement in this direction. The HP TouchSmart and the recent Lexmark announcement of under $400 inkjets that connect to the web.

    Given that most of the manufacturers MFPs – laser, inkjet and solid ink – have links to SaaS it only seems plausible that production equipment will follow. My 2 cents is that as the connection between the cloud and the output devices become deeper and more seamless print will re-emerge as the most mobile information platform. As QR codes are implemented in more and more print products another strong connection will be made.

  2. Jeff Lazerus

    “A related benefit of simplifying the workflow and the VDP development system is in the area of required skill sets. It may be easier and less expensive to find people who can use the basic tools and then grow their expertise, rather than recruiting people with fully matured skill sets that can facilitate the most complex integration workflows walking in the door. People with matured skills are likely to be few in number at premium prices.”

    Are you saying that the low frequency of people with high-level skills in VDP programming is a perceived barrier to adoption? Or is your argument merely for the “simplification” of VDP workflows so that VDP implementers don’t have to employ those with “fully matured skill sets”?

    I would like to see the whole argument in its context. Many of the statements here don’t have any supporting evidence.

  3. Sara Kmiecik

    Interesting, especially the information about the printing and early infrastructure. I will definitely be checking out these books when they come out.

  4. Noel Ward

    Ah, the good old days of line printers, mainframes, and DEC laser printers connected to a VAX. Been there, done that, have a few arrow scars on my back. Life as a pioneer!

    M. Barzelay is right on the need to simplify things for creation of variable data documents. This has to happen at the points where a given document design is merged with data. There are lots of good tools for this, and they keep improving. However, any transactional or direct mail service bureau worth a bottle of black toner can write any code they need to create the VDP app du jour.

    Still, outside of the bureaus or IT departments, we still see the same problems that were evident when VDP for something other than transactional first began to evolve. Many databases were not built in a way that made them accessible or readily usable for emerging VDP marketing or trans-promo apps. This is changing somewhat, but it’s still a significant barrier, and it limits how easily and effectively data can be integrated into documents.

    Then there is the designer issue. Too few designers have a grasp of how data can be used in documents. Many have basic skills but still need training to become more comfortable with data and realize the power it can bring to their design work. This applies not only to graphic designers, but also those who labor in IT departments and bureaus, creating transactional docs that, just maybe, will include trans-promo content. Learning to use the tools for document creation is one thing. Understanding what goes on “under the hood” is another and is ultimately just as important.

    And MJ, I have less faith in the Cloud than you. For some things it will work, but I don’t expect it to be the choice for transactional work anytime soon. Way too much risk to sensitive and/or personal data.

  5. Nick Barzelay

    Jeff, I believe that scarcity of high-level skills is one barrier to expanding VDP beyond mail merge. The real barrier is finding people with cross-disciplinary skills. VDP is difficult because it involves left and right brain activity — one side artistic and the other side engineering. Look at education programs: how many IT related programs teach design tools and skills, and how many design programs teach programming and database tools and skills? The answer is, very few. I’m aware of two: Clemson, CalPoly. RIT is ramping up. I have contacted to some schools in my area (around Rochester, NY), and they have no idea what I’m talking about not do they demonstrate any interest in finding out — and talk is cheap!

    More evidence: Go out and check your mail box. How much VDP is there and what is it. What I found after a year-long informal study is that most VDP is simply mail merge. The few attempts I saw at personalization were very limited. Additionally, small or local business that could really benefit from VDP usually sends static-content marketing mailers or nothing at all.

    This goes for transactional-promotional as well. Where is the promotional portion? The promotional that I usually see has nothing to do with my interests — it’s wasted ink. A credit card has all my purchases. They can extrapolate what I buy and where I buy it. None of that information is used to customize promotions for my bill. If a bill is going to be “promotional”, they should promote their customers’ interests. Besides, they stuff the bill with all kinds of other junk, which defeats both the economic and personalization aspects of trans-promo.

    So it comes down to this: All the wonderful tools out there with their nice graphical user interfaces, cross-media capabilities, interactive event-driven plug-ins to other standard tools, and some involving considerable expense, are essentially being used to produce mail merge for static-content marketing mailers that are probably being run on an offset press. We are essentially using offset workflows to guide end-to-end digital document development and production.

    Sorry if I sound preachy or a little adversarial. Some of my frustration is probably coming out. I appreciate our questions.

  6. Eric Raymond

    I think the conclusions are totally wrong here.

    By simplifying, you’re taking the wind out of the sails that makes VDP powerful.

    This is an ROI-driven outcome-based medium that should only be used for the “Most Growable Customer” segments. All too often we see marketers viewing VDP (and salespeople selling it that way) as an ‘either/or’ proposition vs. less expensive digital printing (shells w/ imprinting).

    The analogy is test driving a Ferrari and a VW Bug (1968 version) and comparing the 2; but limiting the test drive to my church parking lot. The power of VDP is the power of ‘customerization’ which means each marketer needs to map their customer segmentation strategy w/ ceative development and analytics that further engages customers and encourages stronger response to a personalized CTA. Each marketer needs to define their own point of diminishing returns from VDP based on their own KPI’s, and while vertical market #’s might provide some guidance, it’s up to the service organization driving VDP that needs to step up and own the final results….or stick with conventional printing.

    Remember, the role of advocate is to take the complex and make it simple. So, if this is what’s intended by the conclusion, so be it. If you don’t paint a long term picture of what it’s going to look like, however, marketers will only take baby steps re; segmentation, and you’ll never ever get the economics of VDP to win out vs. other media.

  7. Michael J

    Noel,
    re the security of the Cloud. If Google can use the cloud to store medical records and IBM is selling Cloud computing to every global enterprise, I think the security issues are resolved good enough for almost all transpromo or VDP campaigns. Just last week Google announced that they have implemented security sign ins for education enterprises.

    Put that together with some parts of the municipal government of Washington DC signing up with Google Apps at $50/ seat instead of a Microsoft or HP license and I think it adds up to the Cloud being ready for primetime enterprise computing.

    Nick,
    I think the under appreciated fact is that there very few printers for whom it makes to get into this with both feet. Simple applications for simple problems is probably exactly the low hanging fruit and what is practical for them. It’s mostly mail merge on steroids, but there are many many people who just want mail merge on steroids. If a printer and print and address at essentially no extra cost that’s good.

    But the heavy duty analytics that garners the real money is beyond the capability of most printers and most marketers. Data mining is a full time business that requires a real knowledge of statistics, communication theory as well as long experience in the field.
    The reality is that there are very very few CMO’s who have the capability to do effrective data mining.

    Notice that in yesterday’s announcement of the HP + Consolidated deal for National Geographics, National Geographics used a separate vendor to do the GUI, the W2P and presumably the data analytics. As I read the press release it sounded to me that they did not choose either HP or Consolidated to do that part. If HP and/or CGX didn’t get that piece, does it really make sense to be telling printers this is what they have to do?

  8. George F

    As companies downsize there are less employees expected to do more.
    A few years ago less software was available for VDP and it was more complicated to use than the offerings of today’s market.
    We are marketing and selling our VDP capability’s and I hope the trend continues to make the VDP easier to use.

  9. Nick Barzelay

    Eric, From my perspective, there are two kinds of simplification — one is simplifying the product and the other is simplifying tools and techniques (processes). It does not logically follow that simplifying process will simplify the product.

    However, a more complex process may reasonably result in a simplified product because of incidental difficulties and stumbling blocks generated by complexity. Hence, we see a lot of mail merge, when the power of VDP, as you say, is in customization. Highly customized products can be created with simple tools and techniques.

    What I see is a lot of folks driving their Ferraris to the grocery store, so they might as well be driving a VW Bug. If a person has a Ferrari and is too intimidated by it to put the accelerator to the floor, they’ve wasted their money. What I’m saying is get something that will take you to the store and that you can really punch on a track — for instance a beefed up Austin Mini Cooper.

    The response to Michael’s observations and question is similar. I don’t think a person needs to be a knowledge engineer, statistician, and database architect to do sophisticated VDP. An imaginative digital printer or marketer could bring a lot of high power VDP to small and mid-size businesses in their locality by simply looking at their customers’ sales records, repair records, customer lists, store inventory flows, and accounts receivable. It doesn’t have to be complicated in order to reach out with customized communications.

Comments are closed.