Garbage in, garbage out order fulfillment: An obvious solution By Biff Matthews, CardWare International
Garbage in, Garbage out:
Often abbreviated as GIGO, this is a famous computer axiom meaning that if invalid data is entered into a system, the resulting output will also be invalid. Although originally applied to computer software, the axiom holds true for all systems, including, for example, decision-making systems.
Source: www.webopedia.com
he tasks assigned to our industry's fulfillment companies aren't rocket science. They have, however, grown increasingly complex and detailed, while the quality and quantity of data needed for successful fulfillment has remained constant.
Unfortunately, the information provided with equipment orders is often alarmingly inaccurate and incomplete (garbage in). This causes problems to ripple through the industry, increasing costs and frustration at every level (garbage out).
We don't have a crystal ball
At CardWare International, we received an order for three terminals recently. Within the body of the request were line items for one overlay and one quick reference guide. Sounds simple, but it wasn't. We shipped according to the request and were lambasted - twice.
Dressing-down No. 1 was for not "knowing" that three terminals would "obviously" require three overlays. (Sorry, the mind-reading course must have been lost in the mail.)
Dressing-down No. 2 followed shipment of three terminals, each with overlay and accompanying guide. One company representative told us we "should have realized" only one guide was needed because the terminals were "clearly" to be used in the same area. (Sorry, X-ray vision device must have been lost along with the mind-reading course.) That person was then interrupted by a co-worker who said three guides were actually needed. Bickering ensued.
The internal disagreement on the client's end is, unfortunately, the only thing unusual about this incident. It is otherwise a typical example of what occurs when misinformation is provided to fulfillment services.
No one can work efficiently under these conditions. Yet the inevitable response from clients when we mention our concerns is something like, Well, when you see something that might not make sense, why not just call us?
The answer is we would be calling all day long. Our clients demand a low price, and the industry has become lean and automated in response. But bargain prices do not include the cost of chasing down sales reps, data entry clerks or middle managers by phone every time we receive an order containing "something that might not make sense."
Our industry has no resources to spare. For it to work, everyone involved must resist making broad assumptions. When issuing a fulfillment request ISOs, merchant level salespeople and acquiring bank representatives must ferret out all required information. It is no different than setting up a merchant account with a processor - and every bit as important for achieving a successful outcome.
The garbage-in, garbage-out scenario that often plays out due to shoddy information gathering must certainly frustrate merchants. They may start to wonder if anyone in our industry is competent.
Clarity is critical
Car engines in the 1960's were simple mechanical devices that could be serviced by teenage boys. Today, they are complex systems that are far more electronic than mechanical. And few teenage boys - or anyone else - can do anything beyond the most basic service procedures.
Today's business model for boarding merchants is no less complex. It involves the building and downloading of many files. It also requires creating systems to accommodate not only credit cards, but also gift cards, check guarantees, time and attendance records, age verification, and more.
We received another order recently to "build a file for a Hypercom model T7P." Well, on this particular processor, there's one application for thermal and one for dot matrix printing. It's one or the other, and multiple, different capabilities can be in either application.
Yet, even after we called to determine whether the intent was thermal or dot matrix, we were told to "just build an application for a T7P."
Learning is cool
There's a difference between irresponsibility and ignorance. In a quickly evolving industry, ignorance is understandable; you learn. Yet failing to admit shortcomings, combined with an unwillingness to learn, is wholly unworkable.
Fulfillment houses, of course, make errors, too. The better ones acknowledge their mistakes and correct them. But if our actions are not what caused the problem, someone else must pay for the fix. The only alternative is to pay someone more on the front end for exhaustive review and manual follow-up. This is a foolish idea.
Likewise, if the merchant provides inaccurate information, the resulting exception costs should be passed through to the merchant.
Accuracy in, accuracy out
The situation is bad, and getting worse. But fortunately, the solution is clear. People who place orders need to ensure that the information they provide is accurate and complete before submitting it to debit networks, gateways, fulfillment houses or other entities interfacing with merchants.
Yes, this requires greater industry knowledge. But low-tech cars are history, and so is low-tech processing. There's zero chance either will return, and it's incumbent upon those who provide data to fulfill their responsibilities, so the rest of us can fulfill ours.
As third-party administrators, we're held to tight timelines and a high degree of accuracy. We must have accurate information to meet the expectations of the banks, salespeople or ISOs contracting with us.
We can do a lot, but we cannot do other people's jobs, too. There's also a liability issue. And we cannot change the operational and pricing structure within which we all work. It requires lean operation with minimal human intervention.
No future folderol
As automation moves ahead, demands on those gathering, entering and transmitting data on the front end will intensify. Certain fields will be required for submitting orders. Otherwise, errors will flow through the system.
We can build fail-safe procedures. But, returning to the original example, if a request specifies three terminals, one overlay and one quick reference guide, that is what you'll get. And if you don't specify impact or thermal printing, you may not get any printing capability.
Our customer service manager estimates that 80% of orders we receive from ISOs and banks are correct and complete. Of course, that also means 20% of orders have critical data missing. We catch many faulty orders. But in attempting to fulfill them, we are forced to make assumptions and assume a liability, which is not appropriate. (We are not, after all, comic-book-style superheroes.)
Some fulfillment services simply set aside orders with questions, omissions or errors. Volume and service expectations do not allow time for guessing, and they expect that, eventually, the customer will call. This is neither good service nor responsible.
If ISOs want fulfillment companies to establish exception departments, it can be done. But such departments will not be free, and their effectiveness will decrease with each new development in automation. The solution really is to supply careful, complete information at the outset. It truly is that simple. And it truly is critical.
Biff Matthews is President of Thirteen Inc., the parent company of CardWare International, based in Heath, Ohio. He is one of 12 founding members of the Electronic Transactions Association, serving on its board, advisory board and committees. Call him at 740-522-2150 or e-mail him at biff@13-inc.com
|