Should you use existing ecommerce platforms or build a custom one?
Looking into setting up a new web shop for your business or organisation? Not sure whether to go full steam ahead with a ready-made platform or to develop a custom solution? Or something in between? How do you make that choice? The quick answer is that it depends. The key questions are whether your ecommerce offering fits the standard template and which steps of the customer journey you intend to support with the solution.
This blog is part of our book: The Digital Sales Handbook for leaders in IT. Be sure to claim your own free copy of the book.
Pekka Astala, Senior Consultant
A business-minded software developer, Pekka somehow always ends up leading ecommerce projects.
Begin by thinking about the whole customer journey
Before choosing between technology options like Shopify, Magento, Episerver and so on, it is important to stop and think about the whole customer journey.
A customer doesn’t just materialise at the webstore but instead will arrive with a variety of expectations built up during the customer journey that led them there. Before deciding to purchase products, they might expect to find more information, browse more informative or inspirational content, use specific functionalities (like product customisation wizards) or engage in other behaviour, such as interacting with other customers.
Even if you are planning to get started with as simple a solution as possible (which is usually the correct approach!) it is important to ask yourself:
- What is your customers’ job-to-be-done? What does the customer need?
- How will/did the customers learn about the product or service?
- How will the customers end up on the ecommerce platform?
- If from Google, what were they searching for?
- If from advertisements, what are their expectations?
- Are there extra steps required before purchase? Contract creation? Configuration? Location selection?
- How wide is your offering? Are there critical differences between buying patterns, customer segments, internal processes, and data between product lines?
- What are your sales channels? Are there similar requirements between them?
- Customer service
- Field service
- Sales representatives
- All the above?
- What sort of content will help them with their purchase decision? How will you help them find that content once they arrive?
- What other features (such as chats, wizards, etc.) will be necessary to help a member of the target audience to commit to the purchasing decision?
- How will your online and in-store experience need to be integrated?
- How will the product or service be delivered? How about reverse logistics?
- How do the customers expect to be contacted afterwards for support?
- How can you turn the customers into repeat customers? How do you want to stay in touch with them?
- What will the data flows be like? What information needs to flow between the ecommerce solution, ERP, warehouse systems and so on?
- How up to date do you need your data to be (e.g. inventories)? Real-time?
- Last, but definitely not least: How will the web shop be run and managed? How easy should it be? Do you have specified experts whose job is to operate the site?
Asking these questions helps give the right perspective to the web shop technology decision. A common mistake with ecommerce projects is to become super focused on specific technical requirements that make up only a fraction of a fraction of what the customer sees on their journey.
This is what matters: how to ensure that customers find the store, how to convince them to buy the products and how to convince them to return. Usually, the answers have more to do with content creation, advertisement campaign targeting and so on than the ecommerce platform features.
If the exact details of your technical solution aren’t that important, it makes sense to take the easiest and quickest turnkey solution (e.g. Shopify) available and spend your time and effort on the things that matter. While there are small differences in pricing, usability and so on, for most ecommerce needs it genuinely doesn’t matter that much which solution you use: The success or failure of the project will depend on entirely other factors.
Sometimes answering these questions leads to specific needs when it comes to functionality:
- Maybe the products are complex to buy, require a lot of customisation, or have other needs that require rethinking the purchase flow: product recommendation, configuration wizards, etc.
- Maybe the availability, delivery and so on is dependent on multiple factors and requires a lot of back and forth information flow between the ecommerce solution, ERP, warehouse management systems and other systems.
- Maybe the order fulfilment process is unusual and requires a more complicated process than just adding items to the cart and checking out.
In these types of situations, custom development may be a key element in building a service that adequately serves both the end-user and the business’ needs.
How well does your ecommerce fit the standard template?
There are many options for ecommerce solutions. Here are problems they will generally solve:
- They allow the company to manage static content pages.
- They allow the company to manage product information (prices, variants, bundles, etc.).
- They allow the customer to browse products, add them to the cart, pay for them and receive confirmation.
- They allow the customer to register, login and track their purchase history.
If this basic process sounds like your business, there is a good chance you can use a turnkey solution. However, it is surprisingly common for ecommerce needs to differ from that.
You might be selling a simple service with a single product, in which case you don’t need complex product information management, product variants, discount campaigns and so on. In this case, turnkey solutions might still serve your needs well, but you are not taking advantage of 95% of their features. In that case a custom solution — if there are other reasons to consider it — might also be very cost-effective to develop.
Maybe you are selling a recurring service that the customer will want to log in to and manage on a weekly basis. In this case, the business differs from the standard ecommerce needs which turnkey products have been designed to solve, and you will need at least some amount of custom development.
If the way products look, are bought or are delivered is unique for your company, the fundamental processes you or the customers need may be different than those of a standard ecommerce platform. In these cases, it might still be possible to find a way to use existing platforms, but it might require compromises and the platform might end up adding drag as time goes on. Developing a custom solution will allow you to represent your business logic better and allow you more flexibility in customising it in the future.
Another big topic that relates to the decision is integrations. Platforms make it relatively easy to build integrations that fetch product information from one system and send final customer orders into another one. This will still require custom work but usually does not justify a custom solution. However, if there are enough data flows and if they affect different parts of the purchase flow (e.g. more information flows back and forth between the ERP and the ecommerce system when a customer is making the order), it may reach a point where a custom solution is easier to manage than trying to force the existing solutions to work as desired.
Supporting the customer journey
We have already established that the customer journey contains steps both before the user begins browsing products and after they submit their order. It is important to understand how each step is supported and what the role of the ecommerce solution is in the big picture.
A common situation is that there are separate systems used for marketing, remarketing, customer support and so on. This allows you to choose the best available solution for each task instead of a watered-down compromise where one system attempts to do too much. Especially in the beginning, this is almost certainly the correct approach to take.
Sometimes it does make sense for an established and validated business to invest in building a more unified customer experience with a tailor-made solution. This can allow a customer to log in to a single, consistent system for things like:
- Making online purchases
- Requesting support for their purchases
- Managing their warranty and extended support
- Managing their customer loyalty program memberships
- Participating in a community around the product
This usually doesn’t make sense as the first step of a business when everything changes rapidly. But when the business is well-established, such an approach can help differentiate the brand from its competitors. In this case, investing in a solution to serve a larger part of the customer journey can be a brilliant business decision.
Key requirements for any ecommerce solution
Before further discussing custom development, let’s look at the key requirements that any ecommerce solution is expected to meet.
They need to support analytics tracking, A/B testing, responsive layouts that work well on a variety of devices, good search engine visibility, common payment methods, product information management (usually with product variants, product bundles, etc.), discount campaigns, customer-specific order history and … the list goes on.
The good news is that all major existing ecommerce platforms meet these requirements. If they didn’t, they wouldn’t have become widely used solutions. If you can answer “yes” to the question “Is this a really common need for ecommerce?”, the solution — whether Shopify, Magento or something else — either supports it out of the box or there is a good third-party service for it.
A small caveat here is the difference between B2C and B2B store requirements. B2B businesses often have specific requirements (such as customer-specific prices/catalogues, recurring orders, etc.). Often these too are supported, at least as separately installable plugins/apps, but it’s important to confirm these at the beginning.
The bad news is that if you are developing a custom solution from scratch, it is easy to underestimate the scope of the development project.
The costs of a custom development project
There is no single answer for how much a custom development project costs.
Developing the simplest “ecommerce solutions” can be a matter of work-weeks or even work-days: A simple order form and a payment service integration add up to a checkout flow that results in an order confirmation email.
At the more complicated end, supporting complex products, integrations to warehouse management systems for up-to-date quantities, different storefronts for different regions/languages/customer groups, integrating the in-store and online experience and so on might mean that the project can easily take multiple work-years of development time.
This is not an all-or-nothing decision. Instead, there are multiple levels of custom work:
- Theme customisation: At the most basic level is building a new theme to an existing ecommerce solution. These allow you to change the colours, layouts and so on to meet the brand guidelines but also add simple dynamic elements like wizards to guide the user in their decisions.
- Plugins/Apps: Most ecommerce solutions support plugins, which usually allow you to modify the business logic, such as by adding additional steps or logic to the checkout flow.
- APIs & Headless ecommerce: Many ecommerce solutions provide APIs that can be used to synchronise data between them and external systems. These can even allow you to build a completely separate user interface as a custom development project but still taking advantage of the existing solution for business logic, administration and so on.
- Fully custom solution: Of course, it is possible to build everything from scratch. Even then it is usually possible to take advantage of existing services such as a headless CMS service to avoid having to develop content management features, for example.
As a rule of thumb, you should think about the features required to support the customer journey and then select the approach that allows you to add those with the least amount of custom development.
When making these plans, involve developers (whether in-house or through external partners) who have specific expertise in ecommerce solutions. It is a common pitfall for developers to begin a custom solution project for something that could have been solved with a plugin for an existing solution.
Of course, the total cost of ownership (TCO) is not just development, but it usually far outweighs other costs, such as licensing or hosting.
Remember that the development isn’t over when the shop goes live: It is only then that optimisation based on analytics and customer data begins. This will be an ongoing process as long as the business stays active!
While custom development costs more and is usually not preferable if turnkey solutions support the customer journey, a vast number of decisions can be made to manage the cost of the project by choosing the best approach and then managing feature scope within it.
Turnkey product as a steppingstone to a custom solution
There are often two conflicting needs: a quick start with a turnkey solution to validate the business case, and a long-term custom solution to serve the business processes and the customer journey better.
These two approaches do not need to be mutually exclusive.
Companies setting out on their ecommerce journey often have an unfounded fear of duplicate costs as they think of two separate IT projects. But using a turnkey solution to prototype the business can reduce the costs of building the custom solution. The desired final design can be built (first as a custom theme) and iterated on with the turnkey solution. This saves doing all that work with the custom development project.
The users managing product information, for instance, can prototype their processes with the existing solution and see which features of the web shop they use the most. This can save time in a custom solution when the work can be focused on features that are known to be important.
In general, knowing what the business looks like on an existing system and what should be improved can feed into the plans for building a custom project. It can save time, money and effort to proceed in these two steps, even when it is clear from the beginning that the end result will be a custom solution.
Don’t forget your digital sales and marketing
Few web shops fail to meet their business goals due to a lack of technical features. Quite the opposite: They fail more from a lack of effort in digital sales, marketing and promoting. We’ve found that when businesses invest time and effort into something custom, it’s easy to invest hundreds of thousands in it as a development project and be hyper-focused on the development aspects. We’ve seen this in businesses of all sizes — from start-ups all the way to global enterprises.
Whichever approach you choose, make sure everyone involved in the project is focused on the final goal — the success of the business — and not just the development project!
This blog post is part of the Digital Sales Handbook for leaders in IT. The Handbook is a crystallisation of the key themes leaders in IT need to understand in 2021 to push their digital-enabled sales forward. The book includes interviews with industry experts from companies including Stora Enso, SAS, UPM and Tiger of Sweden. Learn how your IT can become an active driver for digital sales!
Originally published at https://www.columbiaroad.com.