Should You Swap To A Headless eCommerce Tech Stack?
There is no question that eCommerce is booming. In fact, according to Forrester, global B2C eCommerce sales will reach $1.92 trillion this year. But with the constant advances in technology, it's important for businesses to keep up and make sure their eCommerce platform is keeping pace. So, should you swap to a headless eCommerce tech stack? Let's take a look.
Lets Pretend Your eCommerce Store looks like this...
You're running Wordpress for a CMS and Woo Commerce for the ecommerce bit. In the past seven years your site acquired three different SEO plugins, another to ensure it was GDPR compliant and changed hosting so it would always be running the latest version of Wordpress. This broke a few of the older plugins, which meant you had to make some updates just to keep the site running.
The site's been redesigned a few times, but it never really feels like a fresh start. It runs OK on mobile, but the Lighthouse score is pretty bad. You don't really want to run the numbers on how many customers you lose to the load times. Moving to another platform is a real pain. Time. Hassle.
None of the decisions which have led to this place were wrong. Until recently, your only way out of this mess would be moving to a brand new infrastructure.
Your Option - Headless eCommerce
As you're considering this move you've heard of 'headless'. Headless in a web sense means that various different services communicate securely with each other and something sits in the middle and collates all of that data together, in order to build your site.
This is the part where I tell you about how going headless is this shining, perfect solution, right? Kinda. I'm not going to lie to you, headless is complicated. But it's complicated in the same way as this:
... takes longer to set up than this:
Let's continue with this metaphor.
You own a stereo, so you can listen to your old records, then a friend gives you a box full of minidiscs. A boombox probably won't play them by default, but you could attach an external minidisc player, I guess. It'll look a bit ugly and means you can't move the boom box around any more, but it'll work.
Adding a minidisc player to a series of stereo separates is easy. Find a spare input on the amp, wire it all up and away you go.
What if something breaks?
In a boombox, if your tape recorder breaks, congratulations! You now have a very heavy radio. If one part of your hi-fi breaks, you can pull out the unit and get a new one.
What if you want to upgrade?
In the boombox, you don't have the option. In a hi-fi, any component can be removed and replaced whenever you like.
Many parts - not just one
You might have different services which handle your CMS, ecommerce ,and product reviews. Perhaps there’s a legacy blog you want to pull in too. Ultimately, all of these services produce and process data. If you can separate them from the front end of your website, they can focus just on the thing they’re good at and not get distracted by the painful job of interacting with users.
Have your cake and eat it
Finetune Partners used a jamstack to generate a website. This is a way of using headless infrastructure to get a best-of-both-worlds approach to your website. Every brochureware page on your site exists as a static HTML file, which means it can load blisteringly fast and is easy for search engines to index. But once the page has loaded, it starts to behave more like a mobile application so the page doesn’t reload any more and the user interactions are seamless.
Just Pay For The Services You Need
Shopify is an excellent ecommerce system, but it isn't really interested in your content. You can expand the capabilities of Shopify with apps but doing so risks straying back into boombox territory: suddenly you need to worry that the app you've chosen will still be maintained in five years time.
It’s not just that individual, dedicated services are better than monolithic services which attempt to do everything. It’s that you only have to pay for the services you need.
Let’s say that your site has a turnover of £10 million, but only needs two content editors. You don’t need to invest in a £1 million CMS. Instead, the CMS part of your site can include just enough features to be useful, for a lower cost. Or a new marketing manager wants to hire ten new content authors. No problem! You move your contract with the CMS up to the next tier and gain access to additional workflow features (or migrate to a different platform).
Security
Using a jamstack approach means that no private data is stored on your server. Only your brochureware pages exist as static HTML files. Everything else will be dynamically loaded in from either your CMS or your ecommerce platform.
This means you present a tiny attack surface to potential malcontents. All other security responsibilities are held by the headless services you use. This is also true for any GDPR concerns - for example storing personal information and payment details.
Speed
Most content management systems focus - understandably - on the content author’s experience. This means that they strive towards having your whole site as instantly editable as a Word document, as long as the current user is an editor.
This is achieved with JavaScript. Lots of JavaScript. This means that for a web developer building your website, there’s a certain payload of code which they are unable to remove and have no control over. In Wordpress, third-party plugins can add to this burden and further bloat the page weight of your site, in a way which only your development team and several thousand customers per month will notice.
Because headless infrastructure separates the front end completely from the back, your development team is free to make the site load as fast as possible. But don’t take their word for it. Run Lighthouse audits and see for yourself.
Flexibility
Going headless also allows you to swap one service out for another. Want to move from Shopify to Centra? No problem! We’ll export the product data from Shopify, import it into Centra and you’re good to go (I might be oversimplifying this a little). The beauty of this approach is that other parts of your workflow are unchanged. Your front end remains (from the user’s perspective) identical. This kind of shift would be impossible using a monolithic CMS which handles all aspects of your internet footprint.
However, this also potentially opens up a (relatively) easy way into the headless ecosystem: Wordpress can run in headless mode. This means you can still use Wordpress as the content management part of your workflow, but rebuild the front end in a more modern way.
The downsides of headless
A headless infrastructure is no doubt complicated. Content needs to be structured. Those structures need to be reflected by the templates. Different data needs to be pulled together and made sense of. This often leads to headless projects costing more than traditional template driven sites.
There is no shortage of competition in headless. This has led to many services offering multiple tiers of payment, depending upon number of users, bandwidth and so on. And it’s not as if traditional monolithic CMSs didn’t come with their own share of hidden costs beyond the licence fees.
Headless requires a specific skill set and a particular way of working. Moving to headless might find you swapping one login for two or more as you and your team need to manage content across multiple platforms. In an ideal world, your headless CMS would be the repository for all of your content - products, contacts, white papers, etc. And while it’s possible to persuade two different headless services to exchange data with each other, for performance reasons, often the best solution is to separate the content.
Is headless your future?
It’s traditional for articles such as this to spell doom for monolithic content management systems, but I don’t see them going anywhere. Finetune Partners needed a tech stack which allowed others to take on the responsibility of managing data and was scalable to meet the needs of our partners as they grew. It isn’t cost effective for us to migrate clients between different platforms as they grow. To be honest, it just isn't our role. Our tech stack choice us was we want to scale and we knew that scaling is difficult on traditional platforms so headless really is built with growth in mind.
