Why Your eCommerce Application Probably Needs To Be Replaced
Username: Name: Membership: Unknown Status: Unknown Private: FALSE
In last week’s Retail Paradox Weekly newsletter, I quoted several recommendations from our most recent benchmark study on the state of eCommerce (Winning At Customer Acquisition In The Digital Shopping Age). Here’s one that I highlighted:
The Legacy E-commerce Platform May Be Wrong For Today
The industry is now arguably entering the third generation of eCommerce. Gen 1 was a standalone channel. Gen 2 was eCommerce integrated with other retail operations. Gen 3.0 is all about E-commerce services being integral to enterprise operations. There are software architectural implications to that new orientation… The technical infrastructure required to reliably deliver a service to the point of need is different than what some retailers may be accustomed to; cloud-based, API-centric, micro-services based.”
As you might imagine, that recommendation raised questions with some readers, for example, “how old is ‘too old’?” and “what ‘architectural implications’ are you referring too?” Let’s try to answer those questions here and now!
How Old Is ‘Too Old’?
eCommerce has been almost entirely driven by the fast pace of consumer technology adoption. Before 2007 (the year that the Apple iPhone was introduced), the eCommerce website produced single digit incremental sales for most retailers. It was really a “nice to have”, but not “essential”. By 2010, all that had changed – driven by the mass acceptance of the iPhone and its rival, the Android. Although it may be hard to remember now, retailers initially thought of “mobile commerce” as a separate channel from the eCommerce website. But that was both shortsighted and short-lived because consumers didn’t see why there should be a difference. In retrospect, it seems obvious – there is no difference, except as dictated by the technology’s form factor. The technology world had an answer for that; in 2010, the concept of “responsive web design” was introduced, and in 2015, HTML5 (which is far more “mobile friendly” than earlier versions of HTML) was adopted.
So, let’s get this all into a timeframe. Prior to 2010 – twelve years ago – digital sales channels were complete and separate from traditional sales channels. But by 2015 – seven years ago – the digital sales channels could support different form factors (desktop, mobile). And that proved to be timely, because around that same time (2014), Amazon introduced Alexa, a voice activated form factor – meaning that now the retail industry at large had to consider the possibility of whole new digital user interfaces coming into play.
One more thing: COVID happened in 2020 – two very long years ago. Because of COVID, retailers rushed to adopt a new concept that they had only been noodling with beforehand: buy-online-pickup-store (or “click & collect”) and its pandemic-era cousin, “buy-online-pickup-curbside”. What that meant is that enterprise customer order management (OMS) capabilities had to be integrated not only into the digital sales channel but the physical sales channel as well. Thus it was that after twelve years of talking about it, “omnichannel” was finally a reality across most of the retail world, to the point where just last week, a reporter asked me if retail today is “omnichannel”, “unified commerce”, or just … “retail”.
So how old is too old? With this timeline as a guide, if a retailer’s website was implemented prior to 2015 – and hasn’t had a major overhaul – it fits right into the “too old” category. If the website isn’t integrated with OMS capabilities, its outdated no matter when it was implemented – omnichannel ordering is now a consumer expectation for almost every retailer.
What ‘Architectural Implications’ Are You Referring To?
This goes well beyond responsive website designs; what we’re referring to is a jumble of new and important software design concepts and how application functionality is delivered to those who need it. I’m referring to a veritable Buzzword Bingo of techno-babble terms like “headless commerce”, “software as a service”, “microservices-based applications”, and “cloud computing”.
Over the years, RSR has tried to demystify every one of these terms by explaining each in plain language, but at its most basic, advances in computer science have made it possible for business rules and data to be delivered to the person who needs them over a network when they are requested. Anyone who uses PayPal experiences this in real time, very likely on a mobile phone. While there’s no intention here to minimize the importance of any of these technical concepts (they are really important! Ask your CTO!), the best way to summarize them all is to say that computer applications are built that way nowadays – and it’s very different from how computer applications were built a decade ago.
So how long have these ideas been around? Let’s take “headless commerce”: that idea came from a 2013 Forrester Research report that made the point that e-commerce vendors were falling behind user experience trends (for example, desktop vs. mobile eCommerce) and recommended “loosely” coupling the “backend” functionality of a website (all the business rules) from the “frontend” (the user interfaces).
How about “software as a service (SaaS)”? That acronym has actually been around since the mid-1980’s, but arguably, SaaS as we know it today came into being when multi-tenant applications (more than one client company using the same instance of a computer application at the same time) could be made available via “the cloud” (more on that in a second). Technology companies started promoting that concept in earnest about 2012.
“Micro-services”? That concept is about how computer applications are coded. The term was agreed upon at a 2012 technology symposium.
“Cloud”? This term has been argued for thirty years, but for retail it became relevant in the early 2010’s, when Microsoft, IBM, and Oracle all began offering pay-as-you-go application, platform, and infrastructure resources over the Internet.
Tying it all together, nowadays “headless” applications, frequently built using “micro-services” constructs, are delivered as “software-as-a-service”, via the “cloud”. If a retailer’s eCommerce systems are not based on these technical underpinnings, the business may be working at a disadvantage.
Why Your eCommerce Application Probably Needs To Be Replaced
In the benchmark mentioned earlier, we noted that in another (June 2021) RSR study, we had asked retailers to tell us how long they have been using their current eCommerce platform (Figure).
Figure: Winners Reaped The Benefits Of eCommerce Early
Source: RSR Research, June 2021
In the new study we observed that:
“…over-performing retailers have been at the eCommerce game longer. But being an early adopter isn’t always an advantage. Technology advancements in the last 5 years have delivered better performance, flexibility, and scalability, and are enabled by an ecosystem of technologies such as microservices, ‘headless … designs, and cloud-native SaaS platforms.
This is important because it underlines how fast available technologies are advancing. If nothing more, the fast pace of innovation in technology strains retailers’ internal IT capabilities. … new SaaS-based solutions are constructed of loosely integrated services, often with a mix of commercial multi-tenant capabilities augmented by open-source and proprietary add-ons that are frequently updated.”
Only the very largest deep-pockets retailers can afford IT staffs that can keep up with the rate of change that the technology world maintains, or with the speed of consumer adoption of all that the technology world offers. Consumers are already experimenting with using eCom services embedded into streaming social apps like Instagram. As new user interfaces are brought to market, it’s entirely feasible that people will want to shop with the new media.
In the January report (and repeated in last week’s Retail Paradox Weekly blog) we concluded that, “while human interfaces will continue to emerge, the rules and data that drive them need to be consistent across all UI’s… <so> the technical infrastructure needs to be designed specifically to enable new UI’s and new functionality in step with consumer expectations.
And that is why we stand by our recommendation: the legacy eCommerce platform may be wrong for today.
 Retail E-commerce In Context: The Next Iteration, RSR, June 2021
 Winning At Customer Acquisition In The Digital Shopping Age, RSR, January 2022