As you consider how best to take your brand headless, you’ll likely encounter the fiery “best-of-breed vs. all-in-one” debate in your research. It’s an important conversation for many reasons, including cost.
To set the stage, “best-of-breed” refers to using different vendors to obtain the best offering in each application area. For example, a company might use a content management system (CMS) from one vendor, and an order management system (OMS) from another. It’s a curated tech stack of the very best tools available. Some of those top applications may even be built in-house if it's a solution that the engineering team thinks will differentiate the brand and offer a competitive advantage.
Conversely, an “all-in-one” or end-to-end solution refers to a traditional, monolithic system provided by one vendor. It offers various solutions within its platform to meet a variety of business needs.
The right tech stack approach is entirely unique to a brand given its size, goals, and growth strategies. That said, when it comes to headless commerce, we feel a best-of-breed, microservices approach is the way to go.
We believe best-of-breed fundamentally supports not only what headless commerce is—a decoupling of the backend and frontend—but the spirit of what it’s intended to do: give you separation, flexibility, and control over your tech stack and webstore.
Some “headless commerce” solutions available have a strong opinion on what your backend components should be, or even encourage merchants to use backend systems from the same vendor. This defies the spirit of what it means to truly be “headless” because now the tech stack components are closely coupled and interdependent on each other.
The conversation around cost only strengthens the argument for best-of-breed.
All-in-one pricing can be attractive, especially for SMB companies with leaner budgets and small-but-mighty teams, but it’s important to have the big picture in mind. For enterprise eCommerce companies, and companies on the verge of moving upmarket, best-of-breed offers unparalleled flexibility and agility that can’t be achieved when you’re locked into a lethargic all-in-one system.
In order to get various aspects of an all-in-one system to work for you as you grow, your engineering team might have to create deep workarounds in order to “hack” the system so it serves your needs. This consumes a lot of valuable time and resources.
For companies that have the transition from monolithic to microservices already on their radar, implementing a headless commerce platform first could make the process much easier and more cost-effective.
From technical debt and ROI, to indirect savings, this blog will explore how a best-of-breed approach is a cost effective answer for merchants who are embracing headless commerce and microservices.
Technical debt and a best-of-breed tech stack
Technical debt is a metaphor for the side effects of software decisions that prove to be costly. Incurring technical debt can be a strategic decision at times, and it’s even sometimes impossible to completely avoid. That said, it needs to be an informed consideration when comparing best-of-breed to all-in-one systems.
There are several types of technical debt. One type is deliberate, as previously mentioned, where business leaders accept that there will be technical debt down the line as a result of a decision. For example, a team might consciously select a poor-quality solution so they can conduct a quick implementation, knowing full well that it’s a short-term solution and will need to be replaced.
Accidental technical debt occurs when a technology decision doesn’t unfold as planned. One example might be attempting to “future-proof” your codebase in a way that doesn’t accurately predict what will be needed in the future.
Another example could be pouring resources into building an application that’s costly to build and maintain, when there’s a performant off-the-shelf version widely available to competitors. The upkeep for a tool that doesn’t differentiate the brand is costly and detracts dev resources from something that will offer a competitive advantage.
On that note, it’s important to remember that building a tool is just the beginning, there are also costs associated with maintaining it. The managed services associated with some technical tools shouldn’t be undervalued while weighing costs.
Bit rot technical debt refers to systems that break down over time. This can happen if a system doesn’t age well, or individuals or agencies don’t pass on tribal knowledge about a system when they leave.
Merchants need flexibility, optionality, and scalability to limit technical debt. That’s exactly what a best-of-breed eCommerce strategy is intended to provide. With best-of-breed tools, merchants can add, subtract, and change tools quickly so their tech stack accurately reflects what they need at any given time. In theory, you should never have to pay for—and maintain—technology that’s not working well for you.
The right headless commerce platform will act as an abstraction layer, ensuring that each of these best-of-breed systems don’t become too interdependent on one another. This enables the agility and flexibility merchants need.
Best-of-breed ROI
The ROI from a successful headless commerce build with a best-of-breed approach can be quite fruitful. Merchants can see an average ROI of 20% to 30%, as many of our merchants have. Even a 5% to 10% ROI would be lucrative, more than offsetting the cost of a headless investment and best-of-breed eCommerce strategy.
The best eCommerce KPIs for gauging success are rooted in page load speed, and include conversation rate, average order value (AOV), and sales.
For instance, one of our eCommerce merchants, Barefoot Dreams, saw a record breaking 64% uptick in conversation rate after launching a headless PWA and increasing site speed by 34.5%. As a result, the Barefoot Dreams team doubled their sales.
Indirect ways to save with best-of-breed
Though it might not be a direct correlation, the benefits of adopting a best-of-breed eCommerce architecture will touch many corners of your business. Freeing up time for your top talent to work on priority projects can ultimately translate into more money for your brand.
People: An optimized tech stack and backend management can impact the headcount required to operate your webstore. The right headless commerce platform in itself simplifies management because it gives you one codebase to manage across devices. It also consolidates APIs from your best-of-breed stack into one.
This means multiple teams won’t be required to maintain multiple codebases, and changes can be made in one place, by one team. Additionally, some best-of-breed solutions and headless commerce platforms offer managed services that would otherwise need to be handled in-house.
Time: What could your team accomplish with more time? A best-of-breed eCommerce architecture can create more time for endeavors that will take your brand to the next level.
For example, improved developer workflow is a top benefit of headless commerce, in part because dev teams can use tools like Git Push, which reduce the likelihood of coding mistakes and overrides. A buttoned up workflow means less headaches and more time and resources for other differentiating projects.
Another example is the many benefits of a headless CMS. Non-technical teams are empowered to make content changes themselves and will require less developer assistance, freeing up even more dev time. There are also automation benefits such as eliminating the need to manually push a sale live at midnight.
Flexibility: As discussed earlier, a best-of-breed microservices approach is inherently flexible. It allows you to quickly shed tools, and the accompanying expenses, at any given time. For contrast, in a monolithic solution your team may only need certain aspects of a tool, but you’re locked into paying for everything, even the parts you don’t use.
Best-of-breed challenges
A best-of-breed eCommerce architecture will present some new challenges not seen with monolithic systems. The biggest issue is that these different tools will not inherently talk to each other and share data. You’ll have to find a way to create cohesion among these tools to make the big picture work.
Even if you tie your best-of-breed tech stack together directly with respective APIs, you could face significant eventual consistency issues, where there’s a lag in the information being communicated between systems. For example, a backend update to your PIM regarding inventory might not immediately be reflected on your website's frontend that customers see.
You’ll need to account for data velocity, at scale. It’s also important to note that not all third party application APIs are created equal, and you may have issues with rate limits, poorly designed APIs, or APIs with slow response times.
Furthermore, without a layer of abstraction, your best-of-breed tools will likely need to feed into each other directly. This impacts the benefits that come from truly independent tools, like quickly being able to change one with affecting the others. When tools are closely dependent on each other and you want to make a change, you’ll need to rewrite code across your backend rather than in one place.
Nacelle’s best-of-breed stance
Not only do we support best-of-breed eCommerce architecture, but our headless commerce platform solves many of the challenges associated with microservices and best-of-breed ideology.
Nacelle acts as the layer of abstraction and induces high data velocity. This makes issues with eventual consistency more tolerable (think seconds and minutes of lag time instead of hours or even days). It also eliminates the need for backend components to be tied together, supporting the agile transformation of your tech stack as needed. Developers get one Nacelle API to work with and one codebase to manage, rather than managing various APIs across your stack.
If your company is thinking about making the jump from a monolithic system to a best-of-breed strategy, implementing a headless commerce platform like Nacelle first will make the transition easier and safer. And because Nacelle is also a managed service, engineers and dev teams will have more time to focus on differentiating projects instead of heavy management.
