Automation: When It Makes Sense (Part One)
To automate or not to automate is not so much the question for today’s enterprises—rather when it makes the most business sense. The rewards of adequately applied and orchestrated automation are undeniable—including increased speed and accuracy, which are driving many enterprises to attempt automation at every turn. According to Mckinsey & Co. research cited in the New York Times, 60 percent of companies with more than one billion in revenue have at least pilot programs underway using robotic automation. The trend of companies rushing to automate shows no signs of slowing with Forrester predicting by 2021 the Robotic Process Automation market will reach 2.9 billion.
For as long as he can remember, Mark Evans who runs the Quality Assurance and Automation divisions at Ascendum Solutions has had the same plaque hanging over his desk. “To err is human, to really foul things up requires a computer.” It’s with this motto in mind Mark cautions all those who view automation as a magic bullet, a panacea for all their business ills.
He warns companies lured by increased speed not to overlook the delicate balance of time, quality, and money—as well as the perilous possibility of losing what makes them unique as a company. “If you’re very careful about the approach and just looking at speed versus the tried and true triangle of time, quality, and money, you will get your time but may negatively impact your cost and quality.” Mark continues, “If you go out and automate everything, do it just to do it, you can end up pulling the intellectual capital that differentiates your company, business, infrastructure from the rest of the world.”
Automation, applied in the right cases, can deliver significant benefits—yet employed incorrectly can increase costs, jeopardize quality and even compromise identity. So, how do you know when your business should consider implementing automation?
- When ratios start to falter- When the ratios of developer to tester start to approach one to one due to time, scope or complexity. The agile ratio is 3-1 optimized developer to tester, waterfall is 5-1 optimized developer to tester, and iterative waterfall is 5-1 optimized developer to tester. When these delivery model ratios approach one to one it is a good time to start exploring an automated solution.
- When costs begin to rise- Increased time in the testing phase can lead to higher overall costs. A notable red flag is when testing costs approach development costs. When the cost to deliver trips the scale into the testing phase an enterprise would be wise to consider automation.
- When time to market needs adjustment- Today’s technology doesn’t have the luxury of time in delivery—companies must get from testing to production to delivery as fast as possible while maintaining quality and cost effectiveness. Unit testing can sometimes fail or become non-existent under these time constraints. Two weeks can quickly become four weeks, which can become six weeks during the testing phases. Besides, production can be bogged down by elongated processes and entrenched archaic systems, further delaying time to market. When all this adds up to more time required for the product to reach the customer’s hands—it’s time to consider automating.
Automation, although not a cure-all for every business malady, can provide an excellent solution for enterprises who apply it to the correct problem. It will undoubtedly play a significant role in the future success of many businesses—especially those that recognize when and where the proper applications exist. If your developer to tester ratios are faltering, costs are rising, and your time to market is lagging, it may be appropriate to consider automation for your enterprise.
Part Two will examine precisely when an enterprise should move forward with automation and what should be automated.