Show all articles

Garbage In, Garbage Out

  • Shane Cook
Garbage In, Garbage Out

“Garbage In…Garbage Out” must be one of the more prolific sayings in all of modern computing. The essence of this saying is the rub between the potential of the information that can be gleaned from the automation of certain processes, and the reality of what is actually available for consumption. Too often in our use of computerized systems, we build the minimum necessary to get the job done; a transaction processed, a support issue resolved, a simple question answered. So, while we begin with goal of getting great reports, and learning what our data tells us about what is happening with our business, we often fall short of realizing those goals for three reasons.

First, it is time consuming (and therefore, expensive) to design the “Inputs” so that the “Outputs” are valuable.

Second, the people who are in roles that understand which “Inputs” are necessary to create “Output” are not the same people who build the systems.

Third, the business climate is so fast-paced and ever-changing that maintaining the “Inputs” so that they retain their value is an on-going challenge for those who are often already over-tasked.

To address the first obstacle, devise a system where the Subject Matter Experts (SMEs) can intuitively and easily participate in the process of identifying the data inputs that will be needed to generate the valuable “Outputs” you seek in relation to reports and information about the business. Start with a clear definition of the “Outputs” e.g., Resolution Rate, Conversation Rate, Disposition Analysis, Initial Contact Reason, Service Levels, Contact Channel.

As for the second obstacle, once the “Output” needs are clearly identified, provide a system whereby the SME can easily Tag the data “Inputs” that are required to create the intended reports. Presented within an intuitive interface where they request this information in the form that is most useful to them (e.g., the ability to perform simple mathematical functions, averages, and sums) will yield positive reactions. It is critical that the SMEs have the ability to not only define, but to systemically create the “Output”. Equally important, the SMEs must have a system whereby the “Inputs” can easily be enforced through validation, restrictions on input types, and pre-defined decision trees. This cuts to the heart of the “Garbage In…Garbage Out” conundrum. By strictly defining, enforcing, and guiding the data “Inputs”, you make it virtually impossible to allow any “Garbage In”. Clearly, that is the greater half of the battle!

To overcome the third obstacle, it is critical to have a system that is simple and easy for the SMEs themselves to modify. Adding a new data element, creating a new validation, limiting a series of choices, re-aligning a decision tree, altering data tagging…all of these activities must be contemplated as daily and weekly activities in order to maintain the health of your information reporting system. In order to make this process as intuitive as possible for the SMEs, special attention must be paid to how data is stored and categorized over multiple “versions” of the system. Making changes to the system must become a “safe environment” for the SMEs; a “I can’t mess this up no matter what…” situation that encourages regular updating, experimentation, and creativity from the users.

In the end, “Garbage In…Garbage Out” can be systemically overcome if you have a system that not only accommodates the people who use the system every day, but also provides them with the control and flexibility to make the data they are generating “their own”. Inevitably, when we own and control the data that goes into our systems, we are much more satisfied with the knowledge and information we get out.

Related Posts