Headline News, Articles & Events Resource
For Your Digital Experience

Syndicated News

Stephan Zoder — Contributor to Forbes.com


Headless commerce, or the decoupling of the front-end (head) brand experience from the less-flashy back-office inventory, pricing, and service components, is all the rage these days.

The reason is that Covid-19 illustrated that quick-on-their-feet organizations can capitalize on rapid economic and taste changes. Decoupling the two major spheres of eCommerce allows brand voices to change their tune, while the back-office still operates like a fulfillment machine. A recent example of how the two worlds interact is Walmart’s attempt to slay the 800-pound Amazon gorilla through partnerships, such as the front-end brand Threadup and its alliance with Shopify to enable other small businesses’ back-office fulfillment.

But even if your front-end web experience is visually appealing and your firm’s back-office operation is fine-tuned, maybe by outsourcing it to Amazon, who in the end really has all the data to serve customers better next time around?

The Beauty of Modularizing Processes

If a product is introduced or just changed in its attribution, say a description or new Instagram rebate, it doesn’t require database or process changes anymore, aka coding. The same applies if a new customer attribute, say eye color, is added for more targeted promotions. The front-end is adjusted as needed (even just by configuration), and it calls an Application Program Interface (API) from which it can pick up the payload (content) to display dynamically for your Google search, Instagram marketplace, or online storefront, no matter if it is on your desktop, iPad or smartphone. If a new retail or co-branding partner needs to be added, as Toys’ R Us recently and foolishly did by outsourcing their back-office fulfillment to Amazon, just load their product catalog, introduce and link a new front-end web experience, and off you go. You are now selling through another channel and likely to another audience set previously unavailable to you. Having all of these components sit in the cloud on top of avoiding hardcoding makes the whole premise very affordable and quick as well from a total-cost-of-ownership perspective.

Many of you may think that this is what most brands and retailers are doing intuitively today, but the reality is much different. Hardwired, aka hard and costly-to-change eCommerce platforms from legacy vendors are still out there, particularly in the B2B space (selling to other businesses). The likes of Adobe and especially Shopify are continuing to put a dent into this paradigm with 97%+ second quarter growth this year. Tomorrow we will see if Shopify’s Q3 earnings call perpetuates this trend. With the addition of players like Wix and BigCommerce’s IPO earlier this year crowding this market, the bespoke and rather inflexible way to grow market and wallet share digitally will slowly disappear.

Modern Shopaholic

The reason why I believe that the headless commerce footprint requires a brain is because it assumes there is no disconnect between the various stages in a B2B and consumer journey. After all, the customer is known and well-understood because all information is easily available and clean.

Wrong! In most instances, there is no consolidated, true and flexible consumer or B2B buyer profile supporting the front and back-end simultaneously. More importantly, it is not easily updated when things change in real-time post-purchase, and the profile requires an immediate update to serve more appropriate content for the next phone call, web click, or chatbot exchange.

BigCommerce’s VP of Marketing, Meghan Stabler, stated in a recent post that half of mobile shoppers spend less than 3 minutes on a site. These precious minutes better be optimized to really add value to the buyer and be spot-on in terms of messaging. Obsolete or blatantly wrong information leaves money on the table and just adds to cost-of-service and cost-of-acquisition.

The front-end of the digital marketing stack is very time-sensitive and contextual. Obviously, it is largely concerned with content marketing concepts, meaning serving engaging and sticky value propositions to keep users on the digital real estate as long as possible and make it easy to buy. This requires personalization and increasingly Ai-based analytics to ensure the user views the appropriate online content. Even if such a stack has a personalization engine, what data is it really leveraging? User behavior, including how a consumer interacts with the site, can then be used to optimize SEO and increasingly social campaigns, but what about post-sale interactions?

Your Enterprise Software Stack Will Vary if You Sell Soap vs. Electronics

If an organization sells consumables; like detergent, paper products, or coffee, which do not require some sort of installation, activation, or service, this may work just fine. Web interactions tracked through cookies stored in your browser cache are connected to understand if a user with device ID or IP address X has been perusing the site before and if he came from a Google paid search or Facebook ad. Any historic repeat visits of the same user can be stitched together using a Customer Data Platform with a device ID if the cache on the smartphone was not deleted. This data is somewhat sufficient to pop up a promotion based on customer segmentation logic to move merchandise.

However, only when the consumer or B2B buyer makes the purchase and checks out will his/her self-identification (bill/ship to address) be captured. Only at this last step, the checkout, you are a “somewhat uniquely-identifiable entity” to a company, and the more you cruise the web and buy, the more predictable spend and purchase motivations will be. This pattern allows firms to refine their consumer segments. I say “somewhat uniquely-identifiable entity” because very often typos, third-party data, and internal system overrides fudge consumer profile (first-party) data as it gets moved and copied.

Even if the front-end process was fine-tuned to run the research and buying steps smoothly, consumer product firms with more sophisticated, increasingly instrumented (IoT) products, like home or office appliances, grapple with back-office integration for a seamless end-to-end customer experience journey. As these firms shift their sales strategy from their retail and wholesale channel to a more controllable, higher-margin direct-to-consumer (D2C) channels, they need to collect and manage more user information. For the first time, they need to govern this data, including CCPA and GDPR in mind, maybe not at a larger scale but at a rapid pace-of-change. I say the “first time” because historically, the retailer kept the front-end checkout profile data. Beyond scale and speed-of-change, the back office data around installation and service are also typically handled in silos. This explains why consumer product firms with complex products typically only learn of the buyer profile if and when a warranty registration takes place and why it is not actionable insight.

The customer experience leader of a large household appliance organization recently told me that only 14% of buyers register their product, and it often takes days or weeks to process it. As if this wouldn’t be challenging enough, the next customer interaction occurs when the appliance breaks and requires service. With an average product life span of 8-12 years and about 20% of these products having some sort of failure in their second year of operation and another 30% in the third-to-fifth year span; months, if not years, go by without any consumer interaction. It appears that buying a household appliance is fairly similar to signing up for life insurance, judging by the lack of consumer interaction past the initial purchase and the average duration of ownership.

This presents a challenge to legacy consumer product firms as their low-touch, low margin business model does not move in lock-step with rapidly changing market conditions, allowing them to pre-emptively offer maintenance service plans, extended warranties, and accessories. It leaves this opportunity to other brick-and-mortar retailers, aftermarket service firms, Amazon and home warranty spin-offs of large insurance carriers. If firms were to capture this opportunity, a smart consumer profile is a must-have on top of a frictionless, integrated back and engaging front-office operation.

But what happens without integrating all these interactions quickly and cleanly? What if the consumer or B2B buyer returns items, has questions about installation, use or performance issues? As firms continue to offer performance or subscription-based revenue models, particularly for B2B buyers, the integration of sensor information (IoT) has to occur in real-time and be unambiguously associated with the right device (product), service entitlements, its owner and maintainer? How can this information be integrated quickly and accurately to ensure that offers are not made to consumers who have just complained about product quality on Facebook, Yelp, or Google or even launched a service request with the company’s call center?

Not One Human Brain, but Many Fish Brains (No Offense)

This multi-touchpoint, multi-departmental and multi-process paradigm needs a brain to synchronize post-purchase follow-on activities, just as our body’s brain steers the central nervous system to operate our limbs within a split-second.

A centrally curated, often called first-party-data repository may be the missing link to tie front and back office interactions together. Again, most of us would believe that this already exists, but it is often not one but a multitude of versions maintained by marketing, loyalty, service, and delivery management departments. These independently-operated brains are like fish brains. They are fine-tuned to react to stimuli they are expecting and primarily care about. They connect to a single-purpose application, a personalization, or campaign tool but are irrelevant to other service-type processes.

Even worse, many organizations outsourced some of these functions to external vendors, which charge handsomely for cleaning up and appending consumer information from third-party-data sources like Epsilon, Nielsen, and Acxiom. These vendors run black-box logic to harmonize consumer information, which cannot be readily explained and often overwrite recent internal updates, e.g., address, company ownership, etc. They are also notoriously slow to implement required changes (see earlier example of product or channel changes) because they serve other companies with similar bespoke solutions forcing each company to adhere to a lengthy ticketing process.

Being charitable; I believe that Wix, BigCommerce, Shopify, and other front as well as back-office focused headless commerce vendors assume that this capability is present in large and small enterprises. I consider this notion already a stretch. The cynical part of my brain thinks they don’t really care when implementing their solutions. Ultimately, they are responsible for their workload, front or back-office. Large and boutique consulting firms develop and figure out how to enable a company to maintain the integrated program.

The lack of a solid first-party data database is an understood gap and certainly an achievable goal. The overall successful roll-out of such a comprehensive process chain requires a step-by-step deployment of data management capabilities, process re-engineering, and enablement of the administrating employees. Organizations must be judicious in integrating this first-party data “brain” into existing and new processes. An all-at-once deployment or a roll-out divorced from business priorities and constraints, as crazy as this sounds, is an all-too-often recurring recipe for failure. Many times it renders a stylish and busy front-end and/or an efficient same-day shipping machine. In turn, it will also perpetuate a continuing high cost-of-service, excessive promotional budgets as well as low customer satisfaction rating. Yet then again, why change if product is flying off the shelves? Amazon is waiting to pounce and eat your lunch next year.

For original content, click here.

Featured Case Studies