It’s been exactly a month since the IRCE show in Chicago. The show was totally amazing this year, very intense with some really interesting conversations. I left the show feeling that the e-commerce world was like the Wild West and that from an integration perspective it reminded me of EDI twenty years ago. For me it was all very much deja vu. I guess I’m an old guy by Internet standards.
What do I mean by that? Well, twenty odd years ago, we started VL to offer integrated EDI solutions for bricks and mortar supply chain vendors. Our primary focus in the early days was retail. Over, the years we expanded into grocery and other areas of supply chain integration that were similar to retail. After twenty years we have seen a lot of different integrations, some great and some really bad. So if you permit me some licence, I think I have an idea of things when I talk integration.
Twenty years ago in the EDI world the common refrain we heard was why should I buy your EDI translator when I can build it myself, and a lot of people did. The downside of that battle was the ridiculous compliance testing regimes that many hubs rolled out to combat the heavy support that many of the custom built solutions created. This reached the height of absurdity when you could only use AS2 products that had been certified by a particular organization. In my mind the bricks and mortar supply chain is still paying for the fall out from the early days. That’s my 5-cent rant. The build vs. buy argument has pretty much died in the EDI world, but it is in full swing in the e-commerce world. In the Ecommerce world the build idea currently holds a lot of sway because of the professional service dollars involved for the agencies that build custom programmed integrations. There is however another angle that didn’t exist in the early days of EDI, the app or plugin.
Apple has been credited with many things over the years and the one that has gained the most traction in the world of e-commerce is the app or plugin. If you think about it, what Apple created with the app concept is brilliant. Its simple, elegant and a fast way to get functionality loaded up easily and quickly. In the e-commerce world apps are all the rage. Go to any e-commerce platform and they all have their app marketplaces and if you don’t have an app you usually don’t get listed as a partner. Don’t get me wrong apps do have their place but they also reflect the immaturity of the space. Just like the plug and play EDI solution. When it comes to integration however apps are great until the business starts to scale and that’s when the devil in the integration details starts to rear it’s head.
As the world of e-commerce becomes more mature, we saw it this year at IRCE, and the integration touch points become more and more complex the plugin or app becomes a hindrance to business growth and scalability. I say that categorically because most apps and plugins are designed to accomplish a specific function. The premise of the integration plugin is that it work across a wide number of customers, so it needs to be relatively generic in both its approach and functionality. The app to the consumer simplifies the underlying intrinsic complexity of integration. So by default the app has to have limited functionality. Why? Because the cost of the app would not bear increased complexity and that is where the app model falls apart.
As web store businesses scale their supply chains inevitably become more complex and by default there are more integration touch points. The more data has to be manipulated in the supply chain the more complexity there is. So if I am a web store owner who is dealing with a 3PL or fulfillment house that is connected to my web store via a two way app and I bring on another 3PL how do I now split my orders? Maybe the new 3PL has its own app or plugin but who acts as the arbitrator to ensure the right orders to go the right 3PL. I have seen this exact scenario in real life and the subsequent mess that was created. What if I want more customization than the app can deliver?
I can load up on examples, in a variety of different areas of the internet retailing supply chain, where we have seen businesses struggle with poor integration strategies. So what is the solution? The first and most important piece is that the web store company has a data usage strategy. How is data going to move between the various touch points? Who is responsible for what and how are you going to accomplish the various integrations? I am constantly amazed at how few people actual think about this aspect of their business until its too late.
Finally once you have the strategy in place, who is going to do the integration and maintain the automated flow? Plugins and apps are great, and certainly custom programming has its place but this is where a hosted integration service comes into play.
The key, to me, is that the customer has an understanding of what they want to accomplish beyond generalities. Do they have a strategy? Do they understand the implications of the approach they want to take? Does that strategy actually take into consideration the capabilities of the their partners? It’s is all together one thing to want to have a high level of integration but if your partners are not capable of serving your integration needs either because they lack the knowledge or the back end systems who do you turn to?
So some takeaways:
1. Develop a data flow diagram of how you see data moving between the various parts of your Internet retailing business.
2. Determine what integration is better, served by a hosted integration solution, an app or custom programming.
3. Plan for scalability. At about 25 orders a day the manual process becomes too much for most Internet retailers. It’s not when you reach that point but before that you need to have the plan.
4. Integration takes time and time is money. So if your supply chain has any complexity to it, with multiple integration touch points the cost will be higher.
5. Plan for the unforeseen. Too many businesses see their business model as static when in fact it is dynamic and constantly changing. To stay relevant you need to constantly review. Just because an app works now doesn’t mean it will when you want to add greater control of functionality in another area.
Take a look at our Build vs. Buy Infographic for some things to ponder.
Download the DIY vs. Custom Data Integration Infographic Here