What is Input Management?

Until recently, I thought that I knew the answer to this question but if I put the search term into Google, just about every “definition” ends up referring to scanners.  Fair enough, I know that the term was first coined as a marketing tool about 10 years ago by a company that produced imaging workflow software, but it’s a term that is still in use today and I think that it’s important to understand what it means.  Is Input Management really all about scanning paper?

If we interpret the term at its most simplistic level, one could define Input Management as “a business process that will efficiently manage all forms of incoming information”.  We could break the process down further using specific terms such as transformation, OCR, extraction, classification (all terms that we’re very familiar with, of course) but, when we’re describing the process at a top-level, I would argue that the two most important parts of any definition are “efficiently manage” and “all forms of incoming information”.

So, why do nearly all current definitions of Input Management refer to scanners?  Simple: because most of a company’s incoming information has historically been paper-based and the most efficient way of dealing with this is by scanning the paper, then processing the resulting image through ever-smarter software.

The business benefits of implementing a scanner-based Input Management system have been well-proven, and few companies would now question their value and return on investment.  However, nearly two decades have elapsed since the term Input Management was first coined – two decades in which there has been a seismic change in the way in which we all communicate with each other. 

What’s different now?

Paper, clearly, is not dead despite all the futuristic predictions of “the paperless office”.  Instead, it has become merely one way in which companies communicate with their customers, largely thanks to the widespread adoption of the Internet and mobile connectivity.  Companies and their customers now Tweet each other; they have profiles on social networks like Facebook and LinkedIn.  Nearly everybody has a camera in their pocket that allows them to e.g. email photos of checks, or movies of motor vehicle accidents to their banks and insurance companies.

In other words, information is coming into companies through multiple channels and yet legacy Input Management systems all too frequently force it all through a single channel, as though it was all paper based.  The old alliance of scanner and image workflow software vendors all appear to be saying the same thing, i.e. “convert the multiple channels of digital information to a single image-based channel, then process it the same way as paper”.

Does this make sense?  Well, referring back to my earlier suggested definition of the term I’d argue that this approach doesn’t “efficiently manage all forms of incoming information”.  It’s certainly not efficient to convert digital text information to an image, push it through Optical Character Recognition software (then manually correct the inevitable errors) and save it back as digital text information.  And how, for example, can you efficiently process a movie file in a single-channel Input Management system?

Common sense would dictate that in order to efficiently process incoming emails and attachments, as well as digital feeds such as Twitter, Facebook and LinkedIn, you need to have a Multi-Channel Input Management system that will process this information in native digital form.  But take heart, I’m not proposing for one minute that you throw-away the investment you made in your legacy single-channel solution for e.g. invoices.  It doesn’t make sense to re-invent the wheel, particularly if you’ve tweaked and refined this system over many years and now achieve a very high level of automation.

A key part of any Multi-Channel Input Management system should be the ability to make automated, intelligent decisions about the most efficient way to process a given input.  Let’s take the example above; if you already have an efficient scanner-based invoice data capture solution in place, consider routing digital invoices through this same process.  The trick is to have a multi-channel solution that can natively analyze the incoming digital source and intelligently recognize it as e.g. an email with an invoice attached, which will automatically be converted to a tiff image and passed to the legacy scanner-based workflow.

Because the digital information contained within the invoice email body text & attachment is natively processed, classification and extraction is a very fast process.  Additionally, native digital processing provides a mechanism by which intelligent replies can automatically be generated, e.g. to acknowledge receipt of the email or request missing information.  Meanwhile the original email and PDF invoice can be compressed & immediately archived along with key index information extracted from the email body, all while the invoice is still being processed through your existing legacy system.

Digital Processing Advantages

Processing “born-digital” data natively in a Multi-Channel Input Management system brings many additional benefits, however.  Take the example above of a company that receives electronic invoices to one mailbox (e.g. invoices@company.com) and job applications to another (e.g. hr@company.com).  The company will probably have workflows set-up in a document management system that manage the routing of resumes and the general hiring process, but implementing a Multi-Channel Input Management system at the point of entry for these documents can enhance these existing DMS workflows.

When an applicant emails the company, typically the email body text will contain valuable information in addition to the attached resume.  Together, they form a useful source of information about the applicant, so they should be processed natively by the Input Management system as opposed to being separated, converted to TIFF and pushed through a legacy paper-based system.

By processing such digital information natively not only is the analysis much faster, with minimal manual input, but fuzzy matches can be quickly made with the internal cloud (previous applicants, job advertisement reference numbers etc.) and external cloud (Twitter, LinkedIn, Facebook etc.).  This additional intelligence can be used to automatically correct errors in the application, such as a misspelt job reference number, identify previous applicants (rejections?) and add information from social media that the HR administrator can quickly digest.

There are also benefits relating to compliance that a Multi-Channel Input Management system brings, which are not easy to implement using legacy paper-based, single-channel systems.  If an email can automatically be classified based on the content of both the email and attachments, it can be automatically indexed and archived along with any automated responses.  Because the email was processed in its native digital form, this means that the Internet headers can also be retained in the archive, thereby providing some guarantee of sender authenticity.

What needs to change?

Do all these advantages of a multi-channel approach mean that we need a new definition for input management?

Perhaps somewhat perversely, I don’t believe that we do.  All Input Management systems are defined by the requirements of an organization and the information it receives, NOT by software or hardware vendors.  If an organization needs to more efficiently handle the volumes of paper it receives every day, the Input Management solution should center on scanners.  If the same organization also receives large volumes of electronic data each day, the Input Management solution should address this data natively.

What’s needed, in my opinion, is a shift in focus by the systems integrators with expertise in Input Management.  Instead of implementing single-channel solutions that will, inevitably, compromise on efficiency and productivity gains, these experts should offer multi-channel solutions that address the multiple inputs that their customers now receive, and not just focus on scanners and paper.

Clearly, Input Management is alive and well; perhaps it just needs to redefine itself a little.

Gary Hodkinson

Business Development Director

Inovoolution GmbH

www.inovoo.com