By Robert Fox, VP Emerging Technologies, Liaison Technologies
Have you ever read the instructions on a chopstick wrapper or the packaging information on a bag of foreign snacks? While all attempts have been made to teach you how to use chopsticks properly or entice you to stuff your face full of “choco balls,” something just isn’t quite right. Verb tenses and spellings are off. Prepositions are glaringly absent. And who knows what to make of the drawings? While most of the time you can decipher the code, clearly something has been lost in translation. Without the right information you, too, are at a loss at what to do. Believe it or not, this scenario plays itself out each day in the tech world when it comes to data transformation for business integration. Unlike our scenario above where you might go a little hungry, businesses are at risk of losing time and money due to poor data translation. The need to effectively and consistently manage data across all segments of the enterprise grows more imperative every day as the volume of data, the sources from which it derives, and the ways in which it can be leveraged, expands at a rapid clip.
Companies who lack a true enterprise data model, lack a clear and consistent understanding of their data formats and specifications across their enterprise. Even those that do still must be able to communicate, interpret and correlate their specifications and their trading partners as well. Even with a well-defined enterprise data model, the challenge to keep it updated, consistent and used across the organization is in and of itself daunting.
Developing a successful, scalable integration strategy across diverse enterprise applications – from B2B and A2A to the cloud – is a critical step. Equally as important is the data transformation component to sow these heterogeneous and disparate systems together, which in itself is an exceptionally complicated step in comprehensive and cohesive data management. Enterprise IT teams need to factor this complexity into their calculus when considering their integration strategy. The reality is that improperly translated data can be costly to an organization in an era when information itself is becoming an important commodity of business capital. Backtracking to fix transformation and integration logic can add time and expense to business critical projects. As a general rule, a change that costs $1 to fix during development costs $10 to fix during quality assurance, and $100 fix once in production. Constant backtracking can also increase the brittleness and complexity of systems that are often plagued by these issues already.
So before you dig in to that delicious plate of lo mein or bite off more than you can chew for your company’s integration strategy, remember the importance of proper translation. Otherwise you might end up asking for a fork and making an origami lizard out of that wrapper.