Flexible technology is the future
Digital commerce is still a technological endeavor. But the way technology is architected has changed. Technology architecture must be flexible and extensible to keep pace with digital commerce capabilities (Figure 6). We asked respondents what types of digital commerce platforms they have implemented and plan to implement.
Traditional monolithic platforms and custom-made digital commerce platforms, which are predesigned, packaged suites. However, platforms enabled with microservices, API-first, cloud-native, and headless (MACH) will overtake as the most commonly implemented architecture in the next two years.
The rise of MACH architectures makes sense given their use of microservices and API-first functionalities. These two approaches allow integration of a range of digital capabilities such as mobile apps, voice, IoT, wearables, and AR/VR — which monolithic systems have not traditionally been designed for until recently.12
Monolithic providers have noticed the value of providing flexibility of digital experiences. Earlier this year, SAP announced its new composable commerce solution that gives clients a traditional monolithic approach and flexibility in digital experiences through SAP’s network of partners as microservices.13 While MACH architectures look on track to pass monolithic platforms, monolithic platforms can keep pace if they provide composable architectures that provide flexibility in digital capabilities.
Figure 6: Companies favor microservices and API-first over monolithic architectures
Source: Infosys Knowledge Institute
However, flexibility on its own does not link to performance. Companies still need to be able to plug in the customizable bits that make the most sense for their business. Through linear regression, we uncovered that customizable architecture that impacts performance. This architecture correlates with a 4% higher likelihood of a company being a top performer.
Digital commerce capabilities are as good as the data driving them. The use of real-time recommendations based on customer activity correlates with a 5% higher likelihood of ranking as a top performer. However, roughly 40% of companies have not started to implement this real-time recommendations. We asked companies about the implementation stages of six digital commerce analytics use cases, and the results are grim (Figure 7).
Less than a fifth of the companies have completed implementation of any use case. More concerning, more 10% of companies have no plans to implement any use case we studied. This gap is concerning because it could create a huge performance deficit in the future.
According to a 2021 study, 24% of the executives viewed their companies as being data-driven, while 39% viewed their organizations to be managing data as an asset.14 Most companies are still figuring out how to implement analytics for digital commerce.
Figure 7: Nearly 40% of companies aren't using digital commerce analytics
There is a clear disconnect between the level of implementation of capabilities and their associated analytics initiatives. 18% of the respondents have implemented personalized offers and pricing, 39% have implemented personalized customer service, and 31% have implemented personalized product or service- specific content. However, only 10% have engaged in analytics to track performance of personalization initiatives.
One of the key shortcomings is a lack of commitment to fundamentals. Companies are pulled into new tooling before they maximize the utility of their current stack. Companies should activate their data, use existing tools, and focusing on the core building blocks of simplified reporting and exhaustive testing.
Even worse, the implementation of analytics for personalization capabilities like customer profiling, marketing, and real-time recommendations is much lower than the implementation of these capabilities. Without data and analytics, companies may as well take shots in the dark. Companies must implement capabilities and analytics initiatives in parallel. Otherwise, the ROI in analytics remains unknown.
Our analysis suggests that analytics for real-time recommendations based on customer activity provides a 5% higher likelihood of having top-tier performance. To deliver real-time recommendations, companies need the right data, coupled with advanced AI tools. And the solution needs to continuously evolve to track changes in parameters related to customer preferences and occasions.
Real-time recommendations also tie to personalization capabilities, making offers, pricing, and content more effective, helping improve business outcomes.
The extraordinary success of applications like ChatGPT has paved a way for organizations to explore endless use cases for generative AI.
The exploration has created tools that range from personal assistants to audio editors digital images creators to code developers. Generative AI’s utility has already influenced digital commerce.
The clothing company Stitch Fix began experimenting with generative AI platforms last year. Stitch Fix is an online clothing service that uses algorithms to recommend clothing based on customer preferences.
Recently, they experimented with generative AI image platform, DALL-E 2, to create visuals of their clothing based on customer preferences.15 Heinz used the same AI platform to generate an ad campaign of ketchup bottles to promote their brand.16
Generative AI use cases are potentially valuable across many business functions. For digital commerce, particularly marketing and personalization, generative AI is already proving its value.
Systems integration connects a digital commerce platform’s information system, including inventory, customers, pricing, and marketing.
Much like data analytics, marketing and AdTech systems were ranked as the lowest priority systems for integration into the digital commerce platform. These systems generate crucial data used to design personalization initiatives.
Similarly, customer data platforms ranked fourth and pricing seventh, even though these systems drive the bulk of analytics initiatives.
It’s surprising that, when online commerce experiences are more important than product, price, and promotion, this is the level of analytics coming from all of the initiatives companies have implemented over the past couple of decades. This clearly shows that there is a significant amount of spend on capabilities without knowing the effectiveness.
Regression analysis of the systems integrated and the outcomes they affect points to three key relationships:
Integrating logistics and fulfillment systems significantly increases the likelihood of (a) improving supply chain and inventory management metrics, (b) reducing costs, and (c) strengthening payment capabilities and infrastructure.
Integrating product information management systems increases the likelihood of (a) increasing footfall or traffic, (b) reducing costs, and (c) strengthening payment capabilities and infrastructure.
Integrating invoice management systems increases the likelihood of (a) increasing footfall or traffic, and (b) strengthening payment capabilities and infrastructure.
We asked respondents to rank the top five business systems critical for integration with their digital commerce platforms.
Inventory management ranks highest by the number of times it is chosen in top five, but performance enhancing integrations are less prioritized (Figure 8).
Figure 8: Companies don’t always prioritize integrating systems that improve performance
These relations reaffirm that businesses must take a holistic view of digital commerce, which includes systems integration.
Every aspect affects the experiences a company delivers and the outcomes it derives, whether directly or indirectly.
