App Development Trends Favor Custom Ecommerce Platforms
-
Application development is changing. Frameworks, services, and even tools make it possible to rapidly create web and mobile applications custom to a business’s needs. These changes mean that it may make sense for some large and mid-sized ecommerce businesses to build their own custom platforms rather than using “off the shelf” ecommerce solutions.
The question of whether to build a software application or buy a solution — or lease, if you will, in the case of software as a service (SaaS) — is common to almost any modern business, including ecommerce. Making the decision to create an application versus purchasing one requires weighing many factors, and at the moment, the aforementioned development trends could be tipping toward building an ecommerce platform rather than using one off of the shelf.
Build-versus-buy Analysis
If you searched on Google or Bing for “build versus buy,” you’d find no shortage of articles intended to help businesses think through the build-versus-buy decision. Often, these articles will include lists of common build-versus-buy pros and cons.
Building your own application should:
- Provide just the features you need;
- Integrate better with your other systems and operations;
- Allow for better third-party integrations;
- Provide better performance (faster loading pages for example);
- Better represent your brand;
- Provide a competitive advantage.
The cons for building your own application are typically:
- More expensive;
- Longer to implement;
- Need a much larger staff.
In the ecommerce industry, retailers almost always choose to buy. These companies add to the list of potential cons, citing things like Payment Card Industry (PCI) standards or the need to comply with accessibility requirements. I’ll address these cons.
Building an Ecommerce Platform Can be Less Expensive
When you are considering the cost of building versus buying an ecommerce platform, it is difficult to compete with SaaS services like Shopify and Volusion that offer solutions ranging from less than $10 per month to just a few hundred.
With this in mind, small ecommerce businesses should almost always select this sort of solution, at least for now. As an ecommerce business grows, becomes more complex, and starts to sell through additional channels, that business will often need to turn to “enterprise” ecommerce platforms.
On the low end, you can license an enterprise ecommerce platform like Magento Enterprise Edition for, perhaps, $15,000 to $20,000 per year. You’ll have integration expenses, maintenance costs, product support fees, hosting bills, and database services to pay, too. So your annual expenses will be, maybe, $40,000 to $45,000 for a solid solution.
Solutions as good as SAP Hybris might cost $50,000 or more to license, and annual expenses could easily reach $100,000 or more.
These enterprise ecommerce solutions are said to cost less than building a proprietary solution because many companies assume they will need to pay a team of developers for several months to create the custom solution in the first place, plus keep more developers on hand to maintain it. But this may not be true.
Consider for a moment the emergence of modern application develop frameworks. These frameworks provide a foundation for almost all of an application’s basic features and functions, so that building your own ecommerce solution is not really starting from scratch. It is, rather, like standing on the shoulders of giants.
Take the case of Laravel, a PHP framework. Laravel is built to make the most common application features easy to implement. Need database connections? It’s pretty much done for you with Laravel. Need user management and authentication? The basics are built in with Laravel. Want a way to email order information or shipping notifications? Laravel provides a simple API for Swift Mailer, plus drivers for SMTP, Mailgun, Mandrill (a MailChimp service), and Amazon Simple Email Service.
If this is not enough of a boost, Laravel has a $99 add-on (a “scaffold,” in code-speak) called Spark that includes subscription billing using Stripe or Braintree. The difference in code between taking monthly or yearly subscriptions and processing single orders is small. With Spark providing the code to take payments, it will be a yawn. And with a service like Braintree or Stripe, PCI compliance is relatively easy, too.
Put simply, it may be possible to build a custom ecommerce platform using Laravel in about the same time it would take to integrate an off-the-shelf enterprise solution. There should not be a large development bill.
Once the site is up and running, there will be no licensing fees. Hosting the site on Amazon Web Services or DigitalOcean could run up to $1,000 per month.
So, if your new Laravel-based ecommerce solution costs $20,000 a year to host, you could spend, perhaps, $20,000 in initial development and still be investing about the same amount (approximately $40,000) as your first year on a low-end enterprise ecommerce platform.
Put simply, it may be possible to build a custom ecommerce platform using Laravel in about the same time it would take to integrate an off-the-shelf enterprise solution.
This is hypothetical. You would want to look carefully at your own situation. But frameworks like Laravel might mean that developing your own solution is less expensive than you think.
Building a Solution Might be Faster to Implement
A second concern about building a solution versus buying one is that the former might take significantly longer to implement. But this too, may not be true.
Just a few years ago, a developer would have needed to write significantly more code to get a custom solution up and running.
Back then, it did not make sense to write your own content management system, because WordPress just worked. It did not make sense to write your own email system, because MailChimp, MailUp, and Constant Contact just worked. And it did not make sense to write your own ecommerce platform because SAP Hybris or IBM WebSphere Commerce just worked. So why invest the time?
But, again, development has become easier and faster. Here’s an example. If you wanted to have interactivity on your ecommerce site a few years ago, you would probably use a JavaScript library like jQuery. As helpful as jQuery was and is, newer JavaScript frameworks have revolutionized application development. For example, Vue.js and its resources extension make data binding (syncing data with the application) and AJAX simple.
What’s more, there are many services that you can integrate with your custom ecommerce platform to further save time.
A brick-and-click retailer in the northwestern United States, as an example, implemented a high-end, enterprise search solution in 2015. The company spent more than $24,000 in fees and took nearly 90 days to get the service up and running. It worked well.
Then in the fall, one of the company’s developers connected a bit of custom code to Algolia search. The result was a solution just as functional as the high-end search, and it took just two hours to implement.
If this same retailer wanted a custom ecommerce platform, it could use Algolia to provide the search.
How long it takes to build your ecommerce platform will depend on your requirements. But building need not take much longer, if any longer, than implementing someone else’s solution.
Custom Ecommerce Platform Might be Easy to Maintain
The next common con for building your own solution is that you will need to hire extra developers and system administrators to maintain it. But, once again, this may not be true.
Infrastructure as a service has become popular. There are a number of tools designed to help you maintain your servers and your code.
Laravel has a service called Forge designed specifically to help a small team or even a single developer host applications on Amazon Web Services, DigitalOcean, and Linode.
IBM’s Compose makes setting up and accessing databases like MongoDB or Redis as easy as using a web app.
In the end, you may need to have a developer or system administrator, but maintaining your custom application does not have to be a huge burden,. Relative to some enterprise ecommerce solutions, it may still be less expensive.
Summing Up
The build-versus-buy analysis is not going away. The balance may soon swing one way or the other. But at the moment, web application developers have many frameworks, services, and tools that make it feasible to build a custom ecommerce platform.
These tools, services, and frameworks may marginalize some ecommerce platforms — those that are overpriced or under performing. It is a golden age for application development. It could be a good time to consider whether building makes sense for your ecommerce operation.