←  Back to overview
Work Talk

How To Set up an Ecommerce Platform in 4 Weeks or Less

Building or overhauling an ecommerce platform and its entire architecture in 4 weeks, how would you go about it? Where do you begin? To illustrate, here's how we did it a while back.

Introduction

Full disclosure, we're not racing against the clock just for bragging rights. Getting that early momentum is part of our development strategy. It's about building and launching only what's necessary to get the answers we need to move forward. Answers to questions that are crucial regarding the product's desirability, feasibility, and viability. This is the premise of agile development, or CI/CD.

I'll be using Pharmabox as my example, the online retail counterpart of COOP Apotheken that we'd rebuilt. If you like to know all about this project, you can check out the case study right here:

In a nutshell, Pharmabox sells over-the-counter (OTC) medicine. It works the same as your everyday online store, but boasts a product catalog of over 45,000 products. We primarily focused on increasing conversions and retention as most of the revenue initially stemmed from one-time customers, and addressed the lack of real-time data management which caused inaccurate inventory and pricing data to be displayed.

Roadmap

Right, so, four weeks. What did we do?

  • 1 week was spent aligning our ecommerce team with the people from COOP, designing a store prototype and testing it with customers during a handful of qualitative interviews.
  • 2 weeks were spent finalizing the design and turning it into code, setting up Shopify and implementing the necessary integrations.
  • 1 week was spent tidying up the back end, pushing small design fixes related to the Shopify theme we were working with, and preparing for the launch of the new platform.

So in reality, we only had two weeks and change to do the bulk of development, what's the secret?

Methodology

Let me introduce you to a concept many of you probably already know, but never heard the name of. It's called composable architecture. Basically, it refers to using best-in-class technology and services from various vendors to build your digital product or service, rather than relying on a single vendor or trying to build everything from scratch yourself.

It's like doing a jigsaw puzzle without having to create the individual pieces yourself. Put it like that and you'd be wondering why everyone isn't taking this route. It's gaining a lot of traction in the ecommerce space specifically, hence it's often referred to as composable commerce as well:

By 2023, Gartner predicts that organizations that have adopted a composable approach will outpace competition by 80% in the speed of new feature implementation.

Source: Composable Commerce Must Be Adopted for the Future of Applications

This ties in nicely to what I mentioned earlier about our development strategy; pushing features and solutions out the door as soon as possible for the sake of learning and growing before your competitors do.

In summary, you take these smaller, pre-exisiting, individual solutions or services that really excel in their specific area, and tie them together. Those specific solutions, that are part of your composable architecture, are also known as best-in-breed solutions. And that's exactly what we did for Pharmabox; using Shopify as the foundation for the store and connecting it to best-in-class integrations. It allows you to piece together your perfect tech stack, but making everything work together in harmony can be challenging.

Pros & Cons

Let's zoom in on the advantages and disadvantages of this approach. Starting with the good stuff:

  • It's easier to adjust or switch around your architecture if necessary, making you more agile when responding to changes in the market or your business needs.
  • There's a lot less risk of vendor lock-in. All-in-one solutions are generally pretty expensive, and takes a lot of time to set up (I.E. Magento, now Adobe Commerce)
  • Faster and therefore cheaper implementation means a higher ROI at an earlier stage. The added agility also creates more security for your company's digital products moving forward.

Although using a composable approach can bring big benefits to your business simply due to the performance potential of each individual solution, there are some disadvantages as well:

  • The potential complexity of integrating different technology solutions to work together seamlessly.
  • You may risk paying for some level of redundancy as individual solutions may overlap in functionality.
  • Not using an all-in-one solution implies that customer support will be a bit fragmented as well.

Use Case

Now, let's turn that theory into practice. Despite Pharmabox' massive catalog, we still opted for Shopify as the backbone for the platform. Getting things up and running was therefore as straightforward as connecting the Shopify store with COOP's own API, which in turn gets most of its product information from Medipim, an online database for medical professionals. To fix the inventory issues, our ecommerce team wrote a new synchronization flow.

That said, we now have Shopify, our blank canvas for web-based shopping experiences. What did we add to make this a best-in-breed solution? What other solutions did we use to fully set up our composable architecture?

  • Shopify Themes: Arguably the biggest time saver when it comes to front end development. We used an existing store theme for the Pharmabox platform, which was built using Shopify's own templating language, called Liquid. While a little cumbersome, this gave us enough flexibility to shape the user experience to our liking
  • Wiser: Remember one of the challenges I mentioned in the beginning, how Pharmabox had too many one-time customers? This is why we integrated Wiser with the Shopify store, dynamic a product recommendation engine that's much more capable than what Shopify itself has to offer.
  • Algolia: Given the size of COOP's product catalog, we needed a dedicated search engine tool to help customers find what they need more easily. And that's exactly what Algolia does, and it integrates with Shopify seamlessly. We also enabled customers to search product based on symptoms instead of product names only; a proper conversion booster.
  • Trustpilot: Social proof plays a large role in getting a company seen and driving its growth. It not only creates trust with customers, it's also a way for Pharmabox to engage with its customers. It's also a great help for SEO visibility as well. Again, we could easily integrate Trustpilot as a Shopify app.
  • Weglot: For most medium to large projects, creating multi-language websites shouldn't be too much of a hassle, and Weglot can handle this very well. We simply connected Weglot to our Shopify store, the app then detected all the content on our site and subsequently generated translations and language subdomains automatically.
  • Mollie: Obviously, Pharmabox customers need to be able to pay for their orders. To do that, we enlisted the help of Mollie: a payment provider for the Benelux, offering payment methods like credit cards, iDEAL, Bancontact, PayPal, SCT, SDD and others. Integration this was as simple as connecting yet another application to our Shopify platform.
  • Sendcloud: Shipping can get frustrating quick in ecommerce land, but integrating Sendcloud with Shopify made this a lot less painful. To make things even more easy, we could simply select the same carrier Pharmabox was already using before, namely DPD. The software also lets you pick, pack, and print shipping labels.

Conclusion

All signs point towards composable architecture being a winning recipe for online shopping applications. The speed at which you can iterate is key to future-proofing any kind of digital experience. For Pharmabox specifically, we've had some great results a few months post-launch. Comparing the data between January and May of last year, we've noted an increased in conversions of 70%. The number of orders placed? More than doubled. The bottom line is that Pharmabox has managed to nearly double its monthly recurring revenue, at the cost of just 4 weeks of teamwork. That alone should make you reconsider this approach.

Join Bothrs

Bundle forces and let’s create impactful experiences together!
See positions