What is testing? (A closer look at data testing) | NodeGraph
What is testing? (A closer look at data testing)
If you know who we are at all - you know that we are huge advocates of data testing. But, before we dive into all the benefits that can be reaped from testing your data, let's go back to basics. In this article, we will be discussing what testing is by exploring different types of testing, diving into the realm of software testing, and explaining its role within big data and business intelligence.
So let's begin by answering the question at the heart of it all - What is testing?
In its most basic form, when we refer to testing, we mean validating elements of your business intelligence solution as a means of assuring high data quality - securing that actual results match expected results. Testing, as opposed to debugging, involves identifying undetected errors, allowing you to always stay one step ahead.
Testing comes in many different shapes and sizes. So many that we could write a whole novel if we decided to explore them all. But we won't, don't worry. Instead, we have focused our attention on some of the most popular types of testing. These include:
Application testing, which is a type of testing that includes validating three separate components involved in the creation of an application. These are data, processes, and output.
So why are we telling you this? Well, both to expand your understanding of data testing and to show you that testing is not a new subject. However, when exploring business intelligence and big data industries, testing is not yet universally adopted. Let's take a (brief) look at the history of testing for one potential explanation to this.
A (brief) look at the history of testing
Testing, or more specifically software testing, originated within the software development industry and dates back to Alan Turing and the development of The Turing Test in 1950. What followed, including international trade and the introduction of the personal computer, increased the demand to ensure quality and thereby the use of software testing. Automated testing soon became the norm - allowing the process to become more air-tight and efficient.
And while there has been a recent downward trend within the implementation of software testing, attributed to the drastically shortened time-to-market requirements, it's still very much in full force. But, interestingly, and rather, unfortunately, testing has not yet become the standard within the business intelligence industry.
Why? Well, we would argue that adoption is still low due to the relative adolescence of the BI industry. This, along with the fact that many BI platforms, Qlik in particular, can be implemented and operated without extensive software development experience - allowing the otherwise standard procedure of testing to be foregone.
However, it's not all bad - we believe that testing is on the upswing. As the size of the big data industry begins to stagnate, the need for quality is superseding the need for quantity and with this -testing will be key. So, please stay tuned to our social media channels the coming weeks as we release more articles discussing how to test, why to test, and testing with NodeGraph!
P.S. A closer look at automated testing
Before we sign off, let's take a closer look at automated testing. Automated testing uses scripted sequences to allow the testing process to be automated. This, in turn, leads to a drastic increase in efficiency as well as accuracy - by eliminating the risk of human error. To find out more about automated testing, click here.
Get started with testing
Do you want to learn more? Then request a personalized demo of our Data Quality Manager today by clicking below.