Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Key Steps to the IoT Epiphany

DZone's Guide to

Key Steps to the IoT Epiphany

Make sure your IoT project is a success.

· IoT Zone ·
Free Resource

Download The Comparative Guide to Rules Engines for IoT to learn what are the most common automation technologies used in the IoT domain for application development

Industrial equipment manufacturers around the world are looking for ways to protect their core businesses while also seeking new cost savings and customer revenue opportunities. Many look to digital transformation and industrial IoT for enabling new services like remote monitoring and predictive maintenance. They're seeking methods for collecting and integrating machine and enterprise data and optimizing their internal operations. Increasingly, they're delivering better business outcomes for their customers through AI and machine learning. For some, the journey is going well. But for many, despite a broad spectrum of technology choices and development approaches, digital transformation initiatives have stalled or ended with connected product systems that drain rather than add to P&L ledgers. To learn which direction your project is headed, ask yourself one question: is your team focused on developing a connected product, or are they building a connected business?

Get Back to Basics

Before Eric Reis and the Lean Startup, and so many other similar "guidebooks," there was Steve Blank and his pioneering work on customer development and its relation to building new products and successful businesses. His discovery, born of years of direct experience and astute observation is simple — get out of the office and talk to your customers. Learn about their needs, challenges, and constraints before committing your team to build any specific product. Discuss and iterate on ideas with potential buyers to learn what they will pay for and how much they will pay you for it. In summary, develop your customers before developing your product. 

But what about Henry Ford's famous line and customers asking for a "faster horse?" First, Henry Ford never said that. Second, Blank is not advocating polling your customers for ideas. Instead, he urges businesses to spend more time up front learning about their problems. Don't guess or assume their pain. Finally, do not ask them to propose a solution. Here's where Ford's declaration, regardless of its true origin, rings true. Yes, their pain may be acute. Their vision, however, is limited. Customers will inevitably ask for incremental improvements. As a result, the ultimate business value of their ideas will be similarly bounded.

Build Your IoT Business Value

You're the innovator. You're the one in a position to understand the market as a whole and create something new. You see from outside the box what your customers cannot. Your first job is to understand your customers' problems. Your second job is to identify possible solutions. But your third, and most important, job is to solve your customers' problems in a manner that creates a sustainable business model for your enterprise.

Launch Your IoT Initiative

Creating an industrial connected product strategy isn't easy. If you've never done it before, it can be paralyzing. We've seen IoT initiatives stuck in this phase for years. You need representatives from across the enterprise. Team leaders from IT, engineering, sales, service, and business managers must all be involved, as each plays a critical role in the commercial success of the solution. Furthermore, assembling the right team is just the beginning. Too often, these committees jump right into debating how they should build a new system before understanding what system to build at all. They start by evaluating IoT platforms and soliciting bids from vendors for custom development. Trials are downloaded. Engineers are told to put together prototypes. Too often, these teams focus on product development first, which as Blank explains, is the exact opposite of how you should build your new business.

Create IoT Value, Not IoT Hype

The problem with applying traditional product development methodology to a new business, Blank continues, is the focus on execution rather than learning and discovery. Milestones are created and deadlines are scheduled for shipping products to customers. What customers? Which problems will your connected products solve? Do your target customers perceive these problems as real pain points? Furthermore, are the individuals who are feeling this pain the same people in charge of purchasing decisions? If not, will the decision makers see the same value in your solution? Finally, how will you reach these decision makers to sell them on the benefits of your connected products? Making a line operator's job easier through automation might be a catchy story or demo at an IoT conference, but if the solution doesn't measurably increase productivity beyond the cost of your new service, it's not going to generate any sales at an industry trade show.

Embrace Your Inner IoT Startup

While today's industrial manufacturing leaders have up to a century of experience designing, developing, and selling new equipment to their customers worldwide, launching a new connected product system presents an entirely new set of challenges. In many ways, the cross-functional team assigned to such initiatives is more like a startup organization than just another project group inside a large company.

This is why Blank's model is appropriate for even the largest global enterprises embarking on the development of industrial IoT systems and connected product solutions. You've got to go back to basics. Find problems worth solving. Establish a product-market fit. Industrial IoT is a domain where, with a nod to Jeff Goldblum, short-term thinking can be fatal. Don't ask your team (or a vendor) right away whether they could build a connected product or service. Instead, first, make sure you do the work and have conversations with customers to confirm whether or not you should build a connected product or service.

Discover and Validate Your IoT Customers

So how does this work? As Blank puts it ever so gently, customer needs and the market can be unpredictable, and "you're going to screw it up." It's an iterative process. First up is customer discovery. Surprisingly, this is as simple as it sounds. Go out into the field and discover your customers. Notice we're not saying to discover your users. You can't build a business on positive reactions. They're obviously a good start, but polite words aren't enough. To validate you've found a potential customer, their words must include pledges of hard currency. Now, you're in the game. You've identified a problem that has real value and found a way to reach the people responsible for purchasing a solution. Critically, you don't need to develop software or build new equipment to determine if there's a business model for connected products that your enterprise can take advantage of.

In some cases, these customers may be your own departments. If your high-margin consumables or replacement part businesses are under attack from knock-offs and service pirates, your high-value problems could come from inside. The bottom line is that you need to keep looking until you've found a problem big enough to affect the bottom line.

Validate Your IoT Solution

Now that you've found the market, the next iterative cycle involves finding the product that fits there. You already have a vision for what your customers need. Here's where you can start putting things together. Start with mockups and storyboards, and then move to scenario walkthroughs. Just build enough to stimulate each round of feedback with minimal resource requirements. Focus on use cases and business outcomes.

This is why evaluating IoT platforms doesn't make sense as a digital transformation starting point. You have to understand the requirements and constraints that the solution must meet before beginning the process for selecting an approach for delivering it.

Prove the Right Concept

Another common misstep early on is spending money or developer time on building an IoT prototype or technology-focused "proof of concept" as a means for showing progress. Data from one of your "things" gets sent to the "Internet" and is displayed via a cloud service on a website or mobile application. There may be applause. Unfortunately, such demos don't prove anything and are a waste of resources. Basic IoT functionality is a solved problem, and following a set of tutorials to connect one machine to the cloud mustn't be mistaken as the first step towards your goal. It's quite literally in the wrong direction.

Never go to internal stakeholders or potential customers with such superficial tech demos. They give unrealistic and dangerous expectations as to how prepared you are to deliver a real system, both to your team and your audience. Stick to testing your business hypothesis. Prove that the solution you're describing to your customers is sufficiently valuable for them to support your proposed business model.

Now, you've found the market and clarified a product that fits. Go ahead and give your engineering team or vendor(s) the green light to determine and propose an approach for delivery. Pushing for tech options prior to business model validation is just "engineering theater." Such requests invite your engineering team to get caught up with shiny objects, as well as encourage vendors to try and sell you something, anything. There are several approaches to industrial IoT platform development, and anyone recommending a specific option prior to the customer discovery process is leading your organization down a dangerous path.

Another benefit of developing your IoT business model before beginning IoT product development is you're able to provide investment guidance to the selection team. By starting with an understanding of expected returns, you can scope an acceptable range for your IoT development and maintenance costs. Cost is irrelevant without a clear understanding of value. Your goal is not to find the lowest cost approach for building an IoT system. What you're looking for is the most cost-effective path for sustainably producing the business outcomes your customers want you to deliver.

When choosing your technology, the concepts you'll want to prove first aren't the easy ones. From your customer validation exercises, you should have a good understanding of where the real challenges will be. Now platforms matter. How customizable and configurable will each option be? What is the process for operationalizing machine learning algorithms and AI so the solution gets smarter over time? What does IoT data management look like, and how can you control the flow to feed your data lake without creating a data swamp? Drag-and-drop UIs may be sexy, but the commercial success of your new connected product line depends more on a flexible architecture, rock-solid data management, and scalable infrastructure.

These are the concepts to test and prove first when you're considering an IoT platform development approach. Once you've chosen the right technology and partner for building your connected products, you can start putting together real presentations for your customers. Showing your customer a system built with demo-ware, meaning something you've cobbled together or used a "click-n-go" service or consumer-focused platform, is one of the worst things you can do for your company. Whether an internal stakeholder or a decision-maker from your target market, showing anyone what appears to be a "working system" that only works in overly simplified scenarios is the wrong move. Predictably, if you've done your customer discovery and validation homework, they're going to want it. They'll want it now. Thus the backpedaling begins. You don't have it. It's just a demo. From an architectural perspective, you have nothing. This is why you want your first working demo to be built on top of the same platform that you'll use in production, rather than throwaway code or something provided by an IoT SaaS vendor.

Companies who make this mistake — of getting buyers excited about connected products they're not prepared to deliver — just contribute to the IoT hype cycle and leave their audience stranded in Gartner's perfectly named "trough of disillusionment." It causes a discontinuity in the sales process. Frequently, this can be fatal. Internal stakeholders lose faith in your team, and customers learn not to trust you, potentially damaging future sales of other products and services. Worse, your customers now see the promise in products that provide data-driven insights for optimizing their operations and may go to your competitors who can provide a production system before you.

Sync IoT Product Development With Customer Development

With your business model validated, and with an initial demo system in hand, your customer development and product teams (whether internal engineering or IoT partner resources) can quickly move to the customer pilot phase. These demos, whether for internal or external customers, should be simple scenarios at first. In stark contrast to the misguided early technology demos that some organizations engage in as part of initial customer discovery, the simplicity is by design rather a necessity. You're not limited by demo-ware technology. Instead, you're just doing the least amount of work required to generate critical user feedback on your system as currently envisioned.

Steve Blank calls this first set of customers "earlyvangelists." They've been identified by the customer development team as having enough pain and sufficient budget to try your solution. This is the group that will let you know whether or not the IoT system your product team is developing is on track. As Blank goes on to emphasize, your goal isn't to gather more feature requests. It's to make sure the functionality you described - remote monitoring, visualizations, notifications, or other optimization services - provide the same level of value to your customers in practice that they did in theory.

Now, real data from real machines are transformed into automated action inside the manufacturing facility (or farm, or laboratory, or wherever your equipment is used), with alerts, analytics, and dashboards in the cloud. This is where you find out how effectively your system reduces unplanned outages and increases productivity inside "earlyvangelist" facilities. If you're unable to make this group happy, you have at least one of two problems. Either your business model isn't as solid as previous conversations led you to believe, or your product stinks. If the former, reboot your customer discovery team and get back to work. If it's a technology problem, then you've either chosen the wrong platform or implementation team, or both. Since you haven't yet built up a large sales and marketing organization for connected products yet, you've got time and budget to review your options and select a better fit. Whatever the reason, the genius in Blank's methodology is in the conservation of organizational resources and customer goodwill as you iterate forward along your digital transformation journey.

Create More Customers and Build Your IoT Business

With the lessons learned from the previous two steps of discovery and validation, the next step is to ensure your IT, engineering, sales, service, and business teams are aligned appropriately for bringing your solution to a broader customer base and supporting the demands of each new service. Again, even for companies with long histories of industrial product development, IoT presents a new set of challenges that your organization must adapt to.

For IT departments, IoT adoption requires changes to your application and system policy frameworks. Suddenly your network surface area extends outside of your enterprise and into your customers' facilities. Much of the value created by your connected product solutions will come from integrating machine data with your ERP, CRM, and other data sources currently managed as independent silos. As a result, IoT accelerates enterprise migration of on-premise applications and data storage to the cloud, fundamentally changing how IT teams operate. You don't have to do it all at once, but you'll need to start planning to support a more connected and data-driven organization.

Engineering teams must take IT considerations into account, ensuring security and scalability, as well as data management and access control, are core considerations of platform infrastructure. Actually, there's no reason for your developers to operate at this level all. Most industrial manufacturers have only a handful of software engineers, and even those who do are likely better served by purchasing IoT infrastructure from public cloud providers and dedicated industrial IoT vendors. These components won't provide any unique differentiating value for your business and create an enormous liability when you try and roll your own. You're building an industrial information company, not an IoT platform business. Set your development team (or experienced partner) loose to innovate on top of a proven infrastructure. Don't chain them to a long development cycle and endless maintenance burden. Your ultimate success depends on using your data for optimizing operations and improving business outcomes more effectively than your competitors. Buy your infrastructure, and build your value.

In many ways, connected product business models have the greatest impact on service and sales teams. Traditionally, most organizations either sell equipment to customers directly or through distributors. Customer relationships are transactional, and hardware is increasingly seen as a commodity in a global economy with more "good enough" options at lower prices. Slim profit margins on equipment sales could be made up for through a healthy consumables business (filtration membranes, welding wire, adhesives, etc) as with razors and blades, but these products are under attack as well from vendors offering cheaper versions. Though often of lower quality, they pose a significant threat when customers have no clear way to gauge performance. Similarly, service contracts and replacement parts are profit centers that are being chipped away at daily by a one-two punch of constant push back from customers trying to minimize costs, and service "pirates" offering scheduled maintenance and knock-off parts at reduced rates.

As you transform your products and related offerings, service and sales teams must adapt as well to capitalize on the new opportunities created by industrial IoT. There's a continuum of course, and there's not one right answer. IoT allows you to change the relationship with your customers from a transactional vendor to a valued productivity partner. Data collected from connected products enable remote monitoring and a shift from time-based maintenance to condition-based and even predictive maintenance. That means you can guarantee uptime and offer customers reliability SLAs rather than rolling trucks on a schedule whether the machines need service or not. This is a win for both sides. You create a dramatically more efficient service delivery operation while eliminating unplanned downtime for your customers.

You can replace parts just before they wear out, maximizing lifespan and minimizing costs for customers without risking an outage. This breaks the pattern we see frequently where, in an effort to reduce maintenance costs, facility operators run machines well past the point of expected life, then suffer catastrophic failure during a critical production run. The end result is a frantic call for emergency support and displeasure on both sides. IoT lets you keep customers' operations running smoothly, making you more valuable to them. As a result, you gain a series of lucrative service contracts and protect your replacement parts channel, creating more value for your enterprise.

While connected products and service offerings create opportunities for both cost savings and new revenue streams, your internal teams and processes will require as much retooling as your equipment. Fortunately, a similar model of iterative learning and phased progress used for customer discovery and validation steps apply to this go-to-market customer creation and company building exercises. Together, they comprise your key steps to the IoT epiphany.

See how CEP engines, stream processing engines, flow based programming engines and other popular rule-based technologies perform against seven IoT-specific criteria.

Topics:
internet of things ,iot ,iot solutions ,digital transformation ,IoT solution ,business ,validation ,concepts

Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}