.  In an environment where consumer behaviour is shifting rapidly and supply chain interruptions are likely to continue for months if not years, retailers need to adjust their operations to suit changing conditions. They must build resilience, deliver superior customer experience and ensure that their value proposition is engaging.  Planned transformation timelines have gone out the window and the truth is that many retailers – even the largest ones – are being pulled by the changing market as much as they are pushing forward with their own innovation programs. When you need to move quickly, it’s even more crucial to run the right tests to make sure you’re not making an expensive foray in the wrong direction.  The best model for testing isn’t actually in the retail sector at all. It’s not even in business. If you really want to see who gets testing right, look to the military. Military organisations are complex logistical players whose operations span continents. If they implement the incorrect systems or procedures, the cost isn’t just counted in dollars – it’s counted in human lives. From a civilian perspective, the amount of testing that the military conducts cops criticism for being fiscally “wasteful”, but this is incorrect. It’s deeply necessary.   This same criticism of testing new procedures, initiatives and processes can happen in retail, and it’s just as wrong there. The simple fact is that you can’t plan a website rebuild, factory set-up or store fit-out in a boardroom or next to a water cooler. Resistance to testing on the grounds that it doesn’t deliver an immediate ROI is short-sighted. Testing isn’t about generating return. It’s there to prevent your organisation sinking resources into something that doesn’t work. The return on investment in testing comes from embarking down the correct route and rather than setting yourself up for a loss-making disaster.  1. Identify the whole problem The truth is that testing does cost money. But implementing without testing can potentially cost a lot, more. So how do you go about it? First, know what problem you’re trying to solve. The whole problem. If you’re launching or rebuilding an e-commerce website, don’t just test whether the website works, test how users experience it, and what can optimise that experience. Test the inbound funnels and the delivery and fulfilment. In a warehouse, test the picking, labeling, ERP and fulfilment.  2. Try before you buy As well as testing holistically, you should also test incrementally. If you’re unsure about whether a particular technology is going to work and don’t want to invest in it, try testing a “lite” version of it – a proof of concept. This is one way to overcome the organisational paralysis that grips people facing the prospect of generating a sunk cost in a potentially redundant or backwards system.  3. Make it meaningful You also need to ensure you test meaningfully. There can be a huge and sometimes subconscious impulse to adjust testing conditions in ways that undermine the validity of the test. Sometimes this is about “rigging” the test to conform to a preconceived hypothesis or have it recommend a preferred course of action. More commonly, it’s about risk-aversion.  4. Go big or go home Retailers wanting to test a new store layout will choose an out-of-the-way location or a low-traffic site so that if the test goes badly, it won’t impact their bottom line. This is a mistake. If the test goes badly, you want it to impact your bottom line. You want a clear and real picture of how your planned changes are going to affect your business. Test where it matters. Test store layouts in your most profitable stores. Test website changes at high-traffic times. Do not make the common mistake of sacrificing research validity on the altar of risk mitigation.  5. Never stop testing Finally, testing doesn’t finish once your solution has been implemented. This might seem counterintuitive, but the only way to know whether the decision that you made on yesterday’s data is optimised for today’s business conditions is to be running a continuous testing track. In the online world, this is fairly standard practice (though it may be worth learning for the many e-commerce newcomers). You break off a small but representative portion of your site traffic and test changes on them before optimising these changes for site-wide rollout.  This kind of practice should also be implemented into your supply chain and logistics management as well. The only way to build a resilient network that can withstand the kind of dramatic demand spikes and supply chain interruptions that have characterised the Covid pandemic is to be testing constantly in the background. You may find problems, but you can also find ways to improve what’s already working well. While you might have hit your ROI target on a project, that doesn’t mean the project is optimal – just that it has passed the benchmark that you set. Why settle for a 20 percent ROI when there are ways of obtaining 25?  A culture of continuous testing and improvement is what sets retail leaders apart from the pack. In the old, pre-Covid world it was a “nice-to-have”. Now it’s a “must-have”.  Craig Padoa is the managing director of Wanzl Australia. A C-suite growth specialist, he has deep expertise leading complex business turnarounds in retail. He has been CEO of fashion retailer Siren and has held senior management roles in Pacific Brands companies as well as consulting as a business turnaround specialist for KPMG and KordaMentha. Craig was born into product. As a child he was fascinated by the machinery in the factory his father worked in producing stonewashed denim jeans. He has made it his mission to help retailers discover their passion and authenticity to drive growth.