IT is at the heart of all retail business processes, and it’s inevitable that these systems will fail at some point. While the specific reasons for the IT outages are difficult to generalize, there are a number of factors that contribute: 1.) antiquated legacy systems that are difficult or expensive to replace, 2.) limited IT budgets, squeezed due to challenging comp-sales results and 3.) reduced IT resources due to head office staff cuts.

One of the issues for some retailers is the fact that systems are, in fact, a patchwork of different technologies and functions added over time. As the demand put on them increases they occasionally fall over. For others, it can be about capacity – which is why a lot of retailers have failures at peak times such as Black Friday.

All the reasons listed are the symptoms, but the cause is definitely as Art describes – lack of quality control. It’s rare that I find a CIO who understands the nuts and bolts of “keeping the lights on” because everyone has become enamored with the need for speed, failing fast, and other ways to race to keep up with the business.

My own opinion, as a former CIO, is that the art/skill is gradually falling into the dustbin of history. Yes, CIOs are supposed to understand the business. They are also supposed to understand THEIR business. That includes testing, security monitoring and pro-active activities, and fall-back plans in the event of outages.

This is not limited to retail. Remember the implementation of ACA? That was an embarrassment because it missed the bases of IT 101 — have a single party in charge.

The underlying infrastructure that supports most store systems, such as the store network, is, unfortunately, an area most retailers fail to invest enough in, leading to these types of outages. Too often retailers address network infrastructure after events such as these rather than looking at this as a preventive investment. The result is that critical apps like POS, CRM, mobile, and any others in the store suddenly run slowly when you least want them to. Our own survey-based research has found that retailers may lose up to 3 percent of annual revenues to slow app performance that leads to outages just like this. At the same time, we found that retailers that invest in the right monitoring and control technologies to prevent these slow-down conditions see a revenue increase up to 6 percent. It’s a significant revenue swing that’s at stake. For every retailer I’ve spoken with suffering from these problems, I’ve spoken to another who understands how quickly an ROI can be achieved with this kind of investment. It amazes me this isn’t considered… Read more »

I believe the root cause of these outages is the complexity created by technology silos. Over the years retailers have created islands of automation that have created a support monster. The solution here is unified commerce, one version of software servicing all channels. Until retailers wake up and embrace this vision we will continue to see these types of outages as it is almost impossible to keep this multi-tiered Frankenstein’s Monster running without more people or less complexity.

Two words:technical debt. Retailers have spent years not spending enough on technology and building a robust future-proof architecture. A presentation I saw by an independent analyst a couple of years ago stated: “If you averaged spending 2 percent or less of revenue on IT during any of the last 20 years, you have been under-investing in IT and it is probably the largest threat to your business today.” (Contact me if you want a copy of the presentation.) But they are not alone, about eight years ago I was talking to a retired friend who had spent her working life in banking. I told her about a retailer I had been working with and how all their systems were badly connected and overall very shaky due to unsupported hardware and software. (The retailer in question had bought multiple hardware spares on eBay to minimize future issues from outdated hardware). She told me that retail was not alone – she had seen evidence of the same for years in banking, even though she was not in… Read more »

When companies build themselves too tightly around a technology which is known to have failures, they subject themselves to those failures.

Retailers need to adopt a different stance relative to technology: Today most seem to develop systems presuming they won’t fail. Instead, learn from the airplane industry. Embracing the truth that there WILL BE failure can make everyone safer — including the retailer’s bottom line.

First of all, these are huge, complex systems, so crashes shouldn’t come as a surprise. Second, they are very visible targets for hackers, and so their integrity is constantly being challenged. And third, from a systems and technology position, retailing is still in a transition period. New applications and reporting functions are being constantly added. New algorithms are being developed every day. The number of AI and IoT possibilities has barely been tapped. And customers, expect these systems to work instantly, perfectly, and certainly without interruption. That’s a lot to ask, and failure is inevitable. And this isn’t even considering the vagaries associated with the legacy power grids these systems depend on. So these “incidents” can’t be avoided, only mitigated. And given the level of demand and innovation, even that is going to remain a stretch goal for years to come.

Planning, funding, and installing technology are huge issues for retailers. So huge that retailers are worn out and stop thinking about the issue once the technology is up and running. However, the job is not finished at that point. Contingency planning for problems and monitoring performance are also critical tasks for long term success. These activities are often thought to be unnecessary when the vendor promises that specific technology will work and the retailer does not want to spend more money. However, the hit to sales and image can be quite expensive as well.

A combination of legacy systems being pushed together because of mergers, plus adding on systems to address omnichannel retailing creates a difficult environment to maintain. In addition, the speed of change is reducing the amount of time available for quality assurance and stability testing.