One of the trend buzzwords of recent years is "Unified Commerce" (E-Commerce Magazine), but what does it actually mean? And isn't it essentially the same as omnichannel (Wikipedia)? What are the implications for technical architecture? These are exactly the questions we are exploring this week. Comsysto Reply has been implementing complex e-commerce solutions for clients across all industries for over 15 years. At the same time, we track and analyze relevant trends in our E-Commerce Lab to ensure we are ready for tomorrow's challenges today. We share our insights here – every Tuesday!

Quick topic shift: What was that about predictive shopping again?

A week ago, I wrote here about Amazon Prime's predictive shopping program and new EU regulations regarding returns. Both were, of course, April Fools' jokes, even if they were just exaggerations of current trends and discussions. Nevertheless, we are not aware of any case where Amazon Prime has delivered unsolicited orders based on predictive shopping. However, there are certainly fraudulent attempts that go in a similar direction (e.g., brushing - explanation by hifi.de). The consumer advice center has summarized well what you need to look out for in such cases.

Back to Unified Commerce and Omnichannel

One thing is clear: anyone who wants to remain competitive in the future cannot avoid omnichannel. This means that an online retailer not only uses multiple sales channels to reach its customers (multichannel) – but also that all sales channels have a consistent brand voice and, at least from the customer's perspective, appear seamless.

The launch of TikTok in Germany this month underscores the importance (see newsletter from a month ago) that this omnichannel capability has for an e-commerce system, allowing businesses to easily expand into additional sales channels. This business agility will become even more important in the future. Mega-trends like social media and AI will certainly not pass us by without a trace over the coming years; rather, they will fundamentally change what modern e-commerce systems look like.

What is the actual problem here?

Even if the brand presence appears seamless across different channels, the underlying data and business process landscape is often fragmented. Very often, there are channel-specific systems for inventory management, customer administration, tracking customer activities, and even customer service.

This often has an unfortunate impact on the customer experience. Customers expect an item they just learned about in a personalized promotion to be in stock. Or they expect to be able to easily exchange an item they bought online for another one in a physical store. Meeting these expectations is only possible if business data from different systems is seamlessly integrated and synchronized in real time.

The buzzword: Unified Commerce

In principle, not everyone defines Unified Commerce the same way. For example, if a payment service provider supports Unified Commerce (e.g., Mollie), it initially just means that you can handle the payment process across all channels using their platform, while still requiring other systems for the remaining processes—very much in the spirit of Composable Commerce.

The situation is different when looking at e-commerce vendors (e.g., New Black). Here, Unified Commerce is often used as a buzzword to address the complexity that frequently arises from Composable Commerce and its best-of-breed approach. This perspective often resonates particularly well with companies where the migration to Composable Commerce has resulted in a "zoo" of solutions over the years, which in turn hinders the very business agility they were hoping to achieve.

How does this fit with Composable Commerce?

We already covered Composable Commerce in more detail a few weeks ago . In principle, it is not a major problem for data consistency if an e-commerce system is composed of different services, as long as data ownership for a specific process step is not fragmented. For instance, there can be one system for user management and customer support, from which other systems can query or modify the necessary data. The important thing is to keep an eye on the size of query cascades, as answering a single user request often requires a significant amount of communication between various systems.

However, this problem area is neither new nor specific to e-commerce, and there are proven solution strategies that have long been established in microservice architectures.

Domain-Driven Design and Event-Driven Architecture

Microservice landscapes are often structured based on (business) domains. This allows the software to evolve in a way that is understandable to everyone in relation to the product and business processes, reducing the likelihood of unpleasant surprises throughout the system's lifecycle. This also ensures that specific parts of business processes are clearly mapped to specific services, meaning data ownership always lies with exactly one service. To identify the correct functional module boundaries (Bounded Context), methods from collaborative modeling (CoMo— here you can find an introduction to the topic) are often used, such as Event Storming, so that the existing business process can be mapped as accurately as possible.

To keep end-user request latency low and prevent request cascades, data replication via events is often used. In this approach, a domain service sends events (e.g., via Kafka) whenever data changes, allowing other systems to update their local copies of the required data in near real time. This method offers a good compromise between system independence, data currency, and the end-user experience.

That’s it for today – here is what comes next.

As you can see, the buzzword Unified Commerce hides a whole host of ideas, with the synchronization and consolidation of business data being the most important aspect in our opinion. A Unified Commerce strategy and a well-implemented Composable Commerce strategy are not mutually exclusive.

In the coming weeks, we will take a closer look at how current e-commerce trends and challenges relate to other key focus areas at Comsysto Reply – whether it be collaborative modeling, software architecture, observability, or the renovation of existing systems. Just as with a good Composable Commerce strategy, the rule is: only by effectively combining the best building blocks will we ultimately be successful in the market!

See you next week!

Christian & the Comsysto Reply e-Commerce Lab

‍