"The Internet will disappear. There will be so many IP addresses, so many devices, sensors, things that you are wearing, things that you are interacting with, that you won't even sense it. It will be part of your presence all the time. Imagine you walk into a room, and the room is dynamic. And with your permission and all of that, you are interacting with the things going on in the room."
- Eric Schmidt, Alphabet Chairman
Introduction
The promise of IoT [1] is that (due to the “peace dividend of the smartphone war” [2]) cheap sensors, processing, and connectivity are enabling all kinds of things to be connected and share data that wasn’t possible before. But how suited is the existing IT paradigm to support this? And how do we think about creating new business opportunities and customer experiences in a computing environment so radically different from what we’ve grown up with?
From pull to push
All of us who have been along for the ride during this digital transformation going on for over 30 years now have gotten so used to the pull mode of operating that it’s practically second nature. It works like this: I enter a URL, I click on a link, I tap an icon, and the computer or smartphone fetches the information I have selected and displays it. This synchronous chain of computing is everywhere: ATMs, cars, home electronics, in addition to computers, the web, and connected devices. But this is changing with IoT.
In the IoT computing era, this model will transition to a push one. Just the right information will be pushed to me based on my context, and that context could be location and time. It could be the tasks I’m performing at that moment, or it could be predictive based on an understanding of my calendar and current weather information (better leave 10 minutes early to beat the downpour headed our way).
In this era there will likely be fewer user interfaces as we are used to them today. Interactions with our things may be a bit more natural, conversational, and intelligent. Since buttons are utilised to signal intent, we won’t need as many of them (digital or physical) in this contextual push paradigm. We already see the early twinkles of this near future, with products like (amazon thing that orders laundry soap) and (predictive maintenance? google now?); although, it’s fair to say that we have quite some ground to cover before all of these sensor enabled and connected things are seamlessly talking to one another on your behalf.
In this era there will likely be fewer user interfaces as we are used to them today. Interactions with our things may be a bit more natural, conversational, and intelligent.
From complicated to complex systems
IoT will also change the way we experience things and the way we organise, construct, and operate them. If my “smart home” stops working is it the manufacturer’s fault, the software connecting them, the network operator, or my wifi manufacturer? If my Amazon Echo mistakenly orders more coffee than I need for my Nespresso machine, is it Nespresso or Amazon that I discuss this with? And how would I even initiate that conversation?
The CX (Customer Experience) for IoT breaks down long-held boundaries and structures. No longer can customers assume that a single brand will transparently manage all of its vendors in order to deliver a monolithic solution, or be able to fix problems with any of the parts of the value chain. This will have an impact on IT structures as well. If my autonomous car isn’t working quite so autonomously, is the problem caused by firmware on the car, network connectivity, a bug introduced by an over the air update, a problem with the mapping data the car retrieves over the air, or is it a GPS problem? Perhaps the computer vision LIDAR the car uses to “see” is faulty. The new requirement for interconnectivity between systems of record and systems of engagement complicates network and security architectures (imagine the number of hacking vectors introduced by the increasingly connected car!).
CX for IoT will require those aging IT systems and approaches (homogenous, hierarchical, and self-contained) to become heterogeneous, networked, fluid, and open-ended.
If my Amazon Echo mistakenly orders more coffee than I need for my Nespresso machine, is it Nespresso or Amazon that I discuss this with?
Emergent
Complex systems arise from nonlinear interactions between their components; which is to say, the whole is greater than the sum of the parts. In fact, the behavior of an emergent system cannot be expressed as a sum of the behavior of its parts. You can’t capture the behaviour of a flock of birds by examining the behaviour of all the individual birds. A complex system like a flock of birds might display emergent, coherent patterns. Those patterns, however, result from the behaviour of components making individual, independent decisions. Each bird within a flock decides for itself how to respond to any given situation.
Complicated systems rely on centralised control and hardwired organisational structures. As a result, they work very efficiently until they break. By contrast, complex systems are sloppy and prone to component failure, yet highly resilient. Their decentralised, fluid structure trades efficiency for resiliency. But how do we design for such complex, emergent systems? You can’t, but you can design for the conditions that allow for the type of experiences you wish to provide for your customers.
Emergent systems present both challenges and opportunities to organisations trying to manage complex socio-technical systems. Organisations that leverage the power of emergence can help companies achieve strategic coherency without sacrificing tactical flexibility. Attempting to apply the old IT way of thinking to most IoT implementations is doomed to fail because the value comes from these deployments gathering data within real-world, complex systems.
Attempting to apply the old IT way of thinking to most IoT implementations is doomed to fail because the value comes from these deployments gathering data within real-world, complex systems.
Self-Assembling Services
Bundled in all of this is a completely changing landscape of what it means for us as both end users and people who collaborate in some aspect of delivering these experiences. UX designers are highly sought after nowadays because as technology has matured and become reliable and commoditised, businesses have pivoted to create value for customers through the unique arrangement and application of technology. To put another way: competing on technology alone is no longer good enough because it’s so easy for your competitors to duplicate it. What’s much harder to duplicate is a friendly and engaging experience that expresses your brand in truly delightful ways. Designing for emergent behavior is impossible to duplicate, because there will always be subtle differences that get amplified into bigger ones over time.
UX professionals are experts at synthesising user research across many potential touch points, framing a problem meaningful for both the business and the end user, and creating coherent user interfaces that unify the brand experience. But this has traditionally been done within the space of complicated systems, not emergent ones.
CX for IoT means that brand and experience coherency won’t be about unifying a few different user interfaces, but will be about the dynamic and self-assembling services that pop up and disappear just when you need them to. These experiences will feel highly personal and very different from today’s standard point-and-click or tap-tap interfaces. It will also give us access to interactions that simply aren’t possible today: buildings with embedded sensors can tell us to grab an umbrella or tell maintenance staff that a lightbulb needs changing in the 7th floor women’s restroom. The opportunities for operational efficiencies in almost any manufacturing and industrial company is quite large, as we can move from optimizing for peaks to optimizing for responsiveness.
On the consumer side of business, it is quite common to utilise personas to create empathy and help the team align around high-frequency use cases. Personas can be thought of as composites of your most likely user or customer. We use them all the time, for a number of reasons, and find them to be a low-cost, high-impact method for facilitating our design process and collaboration with developers. But what happens when we’re not designing for the high-frequency use case but we’re designing for a highly personal and individual experience? What happens when we go from design the “big” unified experience to trying to design atomic level, “micro” services or moments that may behave in unpredictable ways? What happens when the same thing happens in the enterprise and starts changing the way we work?
When real-world objects star interacting with people, things get complicated fast. Already we are overwhelmed with notifications, pop-ups, and indicator noises on our laptops and smartphones. When this scales up to include hundreds of new services and applications that span across all of the real world objects we may encounter, it will be a huge cognitive mess.
CX for IoT means that brand and experience coherency won’t be about unifying a few different user interfaces, but will be about the dynamic and self-assembling services that pop up and disappear just when you need them to.
How to get started creating a coherent IoT CX strategy
Over the next few years, sensors, cloud computing, connected smart devices, and real-time big data analytics will combine to deliver a new layer of connected intelligence that will give birth to this new CX for IoT era. Our interactions with information will be highly contextual and just in time, with great benefits for all of us. This era will enable senior citizens to remain independent far longer than they sometimes can today. Far fewer people will die in traffic accidents due to connected cars, one of the top IoT spaces on the horizon. Connected drones will enhance security for both the company and the home, gather data for better agricultural analytics, and do for urban delivery what TCP/IP did for data traffic. And our clothes will become our own personal trainers who help us monitor our activity, diet, and overall well being by prompting us to make good decisions.
If it seems a bit overwhelming, it is. If it seems far-fetched, it isn’t. This is what’s happening now and all kinds of industries are experimenting with what it means to their business and their customers. To help you, here is my three-step process to get started.
Step 1: Identify your internal change agent
You probably know this person, or you are this person: they are conversant in all the latest technological trends, they have a strong point-of-view on the industry they are in and where they think it is headed, and they are a passionate collaborator with a large network of decision makers, builders, managers, and support personnel. If you’re in such a position, designate them as your “Innovation Specialist” and grant them at least 15% time to devote to this area. Even better, create a program to empower a small team 15% of their time to dedicate to pilot projects in this area for your company.
Step 2: Partner for the first project (or few projects)
Now that you have your change agent(s) in place you will want to partner with someone to facilitate a process for a structured exploration of how this might apply to your business. This is something we do with clients and partners through our DMI Innovation Studio, where we have our own innovation specialists, service designers, UX designers, business designers, data scientists, technical architects, and researchers fostering radical collaboration across a myriad of technical domains and industries. We will bring your innovation specialist into our team as a true collaborator and accelerate their ability to gain traction in framing the opportunity landscape for your company around IoT and personal services. This can range from strategy and partnership recommendations to technology choices to prototyping and envisioning future states.
Step 3: Develop a service design capability
Connected products/hardware is a fundamental shift for us from complicated systems to complex/emergent ones, which require a new frame of reference for value creation and design/development of these systems. When we can design a “thing” that can now sense the state of the world around it, convert that into a digital signal that can then be used to take action by something else, then this as fundamental a change to our world as the steam engine was. [3]
3D printing and the low-cost of computation, bandwidth, and sensors means we can prototype new ideas very quickly and test them out, while IoT deployments are increasingly moving to free spectrum (e.g., LoRa, whitespaces, etc.) in order to gain efficiences in battery life and network optimization. This means you can try out lots of different ideas without the overhead of dealing with network operators or traditional manufacturers.
But how do you get any of this into the hands of actual customers and users? The old way of thinking is we pour money into creating a widget and then sell as many of those widgets as possible. This doesn’t really work in this new paradigm though. The “widgets” in this case may often be fairly low-cost, commodity technology. The value they’re creating is in the data and actions that can be taken from that data (automated or otherwise). Therefore, rethink what you’re doing as a service. Abstract out the complexity of managing your network of “things” (deployment, provisioning, monitoring, maintenance, reporting, etc.) so that what is exposed to your customer is only the interesting part for them. Think about how you design for asynchronous, emergent experiences instead of discrete interfaces. Consider the variety of feedback loops and how your software will be continuously deployed based on real-world usage and learnings.
Designing services is a very different skill set than design a product, and many services fail because they are designed with this traditional IT product mindset. Don’t make that mistake!
Step 4: Continue to experiment and test with customers
A key benefit of having a designated Innovation Specialist for your company or department is that there is a clear owner of the outputs from this work. All too often knowledge is created, deliverables are delivered, and it goes on to rest for eternity because there are no clear owners for the output. To mitigate this, we insist on having clear owners and work with them to understand how to formulate our knowledge work in the most impactful and useful ways for them. Sometimes this is a detailed, analytical report; other times it’s a prototype or a video. It could be code and a technical proof-of-concept. What we aim for is to capture the knowledge creation in the most translatable way possible for that particular corporate culture. Afterwards, your Innovation Specialist will need to identify opportunities for pushing the work forward which is where their personal brand and internal network goes a long way.
Summary
IoT doesn’t begin to capture the changes that are coming to our lives because of this new computing era. Our future interactions with information and digital services will be contextual, conversational, and span across a multitude of differing brands, providers, sensors, objects, and interfaces. To understand how this will affect your company and your industry, create an internal innovation program (even if it’s just one person for 15% of their time) as a sandbox to begin experimenting with these new computing scenarios. Seriously consider engaging a partner with a mature innovation process and team with multidisciplinary skill-sets to get you started and accelerate your learnings, while also creating a higher probability for those all important “early wins”. This will demonstrate leadership and build confidence with your internal stakeholders so that you can create momentum for continuing the journey.
[1] Anyone else dislike the name “Internet of Things” (IoT)? It’s neither the “internet” or the “things” that is truly interesting and disruptive about this new computing platform.
[2] “Epiphanies from Chris Anderson”, Foreign Policy, 29 April 2013. http://foreignpolicy.com/2013/04/29/epiphanies-from-chris-anderson/
[3] I highly recommend the book “Smart Things” by Mike Kuniavsky, who originally made this connection between the revolution of the steam engine and our sensor-connected world.
