What is “the web” these days? And what is “the internet”? Wikipedia has a clear definition: “The World Wide Web (WWW), also called the Web, is an information space where documents and other web resources are identified by Uniform Resource Locators (URLs), interlinked by hypertext links, and accessible via the Internet.” For example, HTTP is a web protocol used by web browsers to talk to web servers whereas TCP/IP is a lower level internet protocol for connecting machines. HTTP is built on top of TCP/IP. But where does commerce fit in?

If you have not guessed by now, this is not going to be a deep technical blog. I have been exploring various Google technologies relevant to e-commerce and came across the following question. Is there a difference between web e-commerce and internet e-commerce? And does it matter? I share on various Google technologies in other posts, but the following discussion is a more general vendor neutral discussion.

Clearly there is web commerce. Fire up your web browser, hit a search engine to find a product, go to your favorite marketplace – people are doing e-commerce on the web every day and there are a range of e-commerce platforms ready to help you get online.

But what about smartphone apps? Are apps part of “the web”? No, although they frequently use Web APIs behind the scenes. That is, they use the HTTP to transport API calls to backend systems (e.g. with JSON encoded payloads). Reusing HTTP is convenient because you can share lots of backend infrastructure with your user-based web traffic (firewalls, load balancers, monitoring, etc). The more you can restrict the number of inbound ports and protocols, the easier your systems are to protect from cyber attacks.

But most users (as distinct from developers using APIs) think about the web in terms of starting a journey from a web browser. So apps are not “the web”, although they do use internet (and even web) protocols behind the scenes.

This is where Progressive Web Apps (PWA) get more interesting in terms of blurring the lines. A PWA can do many things a native app can do. It can be added to your home screen. It can send you push notifications even when you are not on the site. It can work even when you don’t have a network connection. Yet PWAs still run within a web browser. (For the nit pickers, yes PWA really refers to a range of “progressive” web technologies that websites can adopt.)

PWA (as well as other technologies such as AR and VR) show a common trend. New technologies are frequently first launched on a device and only accessible via native APIs. Then if the technology takes off, it gets standardized and incorporated into web standards. One clear developer benefit of the web is you only need to develop one application to run on multiple platforms. But this is only true if enough of the major browsers agree on and implement a standard.

Internet of Things (IoT) devices are also interesting here. It is not called the “Web of Things” because users do not interact with a web browser, even if the device uses a web protocol behind the scenes. There is a lot of IoT work these days focussed more on industrial applications, such as remote monitoring of factory machinery. For consumers, IoT includes devices such as voice controlled smart speakers and appliances from TVs to washing machines.

Coming back to e-commerce, is “internet commerce” a thing? Most e-commerce platforms today allow you to create a web presence – your online web storefront. But the good ones also make a robust set of web APIs also available. Should merchants be concerned with hooking IoT devices up to these APIs?

In some cases this may be worthwhile, for special purposes such as in-store kiosks – a dedicated device that a merchant creates and so needs to hook up to their own backend.

But what about smart speaker integrations? Should merchants look for e-commerce platforms with direct integrations to such devices? I think the current answer for the majority of merchants is no. Developing for a smart device today is similar to building a native application on a smartphone. There is no equivalent to web standards across such devices, and I believe it will be years yet before this eventuates. If a merchant wants to build a smart speaker application in the short term (for example, a customer support application, or a customized marketing experience) they need to go “native” and build a solution for each smart speaker vendor (just like a smartphone app). That leads me to the opinion that there is not yet a strong need for an e-commerce platform to support custom applications on such devices beyond making good web APIs available.

What merchants can look for today, however, is platforms that have integrations (either built in or via extensions) with the companies behind the smart speaker devices. The main smart speaker vendors today have commerce experiences for these devices. Integrating with vendor backend services allows merchants to hook into the experiences the smart speaker vendors are creating. This may involve uploading product, pricing, and inventory data to the smart speaker vendor, which is typically easier than building and maintaining a custom application (and hooks in with other services the vendor also provides).

So I do believe “internet commerce” is going to be increasingly important for merchants, where I use the term to mean commerce over the internet where a web browser is not present. Merchants need to be present where the customer is, and given the rate of smart speaker sales (which frankly still astounds me), it is clear that merchants should be thinking about their strategy. But that strategy might not be to build smart speaker commerce experiences directly. For the majority of merchants the best path forward may instead be, at least in the short term, integration with smart speaker vendors. And I think this will be a consistent pattern for IoT devices in general – let the device vendor incur as much of the cost of building out the first experiences as possible, letting the field mature and standardize.