Skip to main content

AI and RPA powered testing: innovate with confidence in workforce management

Artificial Intelligence
(Image credit: Image Credit: Razum / Shutterstock)

When it comes to workforce management, AI is a tool already routinely deployed across all types of companies, in all sorts of industries as organizations look to increase efficiency and agility in their operations – while simultaneously freeing staff from monotonous, low-value work, enabling them to focus on customer service and revenue-generating activities.

Be it powerful algorithms that claim to sift easily through CVs to enable recruiting teams to focus on the best-fit candidates, automated scheduling rotas working across multiple role types, departments and even store locations; or even forecasting tools that claim to predict required staffing levels based on footfall and customer buying patterns, AI-powered tools and technologies now boast a dizzying array of capabilities. 

For busy organizations looking to improve efficiencies, reduce costs and boost employee engagement it isn’t hard therefore to spot the appeal of artificial intelligence in the retail workforce management space. These cutting-edge innovations promise sweeping solutions too good to overlook after all. One global survey of 34,000 workers found two-thirds (64 percent) reported reduced stress levels and a more manageable workload thanks to the introduction of AI, for instance. 

But tempting as it might be, it’s critical that organizations don’t simply leap in blind as poorly implemented solutions can be incredibly disruptive. 

Increasingly, these developments in workforce management tools mean that we’re handing over responsibility for a complex and business-critical system to yet another system. In few sectors is that riskier either than in retail, where workforces combine multiple functions, schedules and demands. But that also makes it more important than ever that in retail companies test the emergent behaviors of software they’re looking to rely upon. Or, in short test these tools to ensure they deliver true business value, rather than simply create disruption. 

For many years, of course, the challenge has been reliance on the ‘old’ ways of testing. Too often organizations are reliant on manual testers, forcing a compromise between quality, cost and time. Test quickly and cheaply and you risk low quality. If you invest in high quality at a low cost it can eat into hours and hours of time. Not to mention that the huge reliance placed on ‘people power.’

In fact, the drawbacks of this way of testing are almost too many to list. 

Challenge and solution

But, they include:   

  • The creation of regression testing scripts that take significant amounts of time and money 
  • The repetitive nature of the work which means mistakes are often made, and can often be overlooked too 
  • Limited resources that mean fixes can’t be implemented in tandem with the testing – adding further time and repetition to the process 
  • A failure to provide sufficient scale or enable load testing due to the small number of scenarios or transactions that can be carried out at one time 

Having such an important piece of work suck up so much manual labor, without even the guarantee of an accurate end result, has left many retailers crying out for a solution for years. 

Thankfully, as much as AI has created the challenge, it has also presented the solution. In particular, by using automated software testing instead, it’s possible to drastically reduce the time and effort involved, and bump up the quality significantly at the same time. 

In today’s fast-paced software testing industry, the need to test and deliver applications quicker and better without compromising on quality is critical to the success of any organization. Companies need to get their products and software updates out quickly but still with excellent quality. An automation first approach enables them to reduce testing timescales – and therefore time to market – while maintaining resource availability and reducing costs for customers.

It’s why in many industries automation testing is already the default option. But in retail, the complexity of technology and systems, and the people-centric nature of workforce management solutions, has meant its adoption has been far slower. 

That could be about to change though. It’s possible to leverage the same new developments that are driving the revolution in workforce management, to create automated testing that reflects the particular demands of retail. That means using RPA processes that allow us to reduce the time it takes for each test by a factor of 400-500x. Whilst at the same time, massively increasing repeatability.

Stepping up with testing

This approach makes it possible to execute around 35 tests of an automated workforce schedule, with full end-to-end validation (including shifts, punches, holidays and timetables), in just a single minute. With a manual scenario tester, the same process would likely have taken them days. In fact, using this same approach it would be possible to automate a set of scripts involving around 1,000 test scenarios just overnight. Manually, it would take six weeks. 

This isn’t just about the speed though. This approach also delivers unwavering accuracy, deploying underlying technology and the ability to test across a wide selection of workforce management activities. It allows developers, business users and stakeholders to identify and resolve defects faster and earlier in the software SDLC process and make decisions based on real-time data. 

For any organization looking to quickly and accurately validate the effectiveness of configuration changes or new solution deployment this shortening of the process is invaluable as retailers need to get products and software updates out quickly but still with excellent quality and unparalleled levels of transparency.. An automation first approach enables the reduction in testing timescales, and therefore time to deployment, while maintaining resources available. Essentially, test strategy planning, comprehensive documentation and extensive automation testing minimizes delays, reduces costs and gets issues resolved sooner.

The reality is as workforce management software gets more innovative, automated and data-driven, so too does the testing of these solutions need to step up a gear, delivering equivalent speed, accuracy and rigor. Until now the inherent complexity of a retail workforce has left this hard to achieve. Not only is it now possible, but this ‘new’ way of testing is more efficient, more cost-effective and of higher quality. All this leaves organizations free to enjoy the fruits of AI, without coming across any bad apples.

Antony Kaplan, Test Services Director, REPL Group (opens in new tab)

Antony Kaplan is the Test Services Director at REPL Group, part of Accenture.