Getting ready for implementing an Industrial IoT project for a customer
Industrial IoT projects can cost millions to buy or develop and still not improve business outcomes. How does this happen to even experienced industrial manufacturers? The details vary. Yet the root causes are the same. A post-mortem of each IoT failure inevitably reveals a common set of fatal flaws. That’s good news. It’s not that the journey is too hard. Many teams just choose the wrong path. Here are the steps taken by those who succeed.
Sometimes companies will diligently identify challenges, create a project plan, and deliver on their goals. Yet the solutions fail to increase their bottom line. Why? In some cases, they deliver a $1,000,000 engineering fix to a $100,000 people problem. Other times, they build a system solving a critical issue or creating a new business opportunity but find themselves painted into a technological corner. They can’t adapt their solution architecture to serve as the foundation for a larger enterprise system, and it becomes an isolated tool. Usually, in these cases it’s better to start over than to continue stacking features atop a shaky platform. Start with a vision for digital transformation across your enterprise, then work backward from there to build your roadmap.
Frequently, the problem or opportunity is plenty big. A chance to prevent hundreds of thousands of dollars lost to unexpected downtime. Eliminate 85% of waste and rework. Create a competitive advantage for achieving rapid market domination. The business outcome is often clear. But your proposed solution? Often it’s a plausible theory – at best. Test it out first. Making a big investment right out of the gate isn’t always the best play. Sometimes bold headlines end with painful retractions. Still, there are many successful digital transformations resulting in business expansion, market dominance, and career promotions. What the difference? It’s said that fortune favors the bold. More accurately, fortune favors the bold who plan ahead.
You’ve identified a business outcome worth pursuing. That’s step one. Step two is pausing to reflect upon the complexity of your challenge. This isn’t just connecting things to other things or showing data in the cloud. It’s not just data modeling and algorithms. While alerts and dashboards are important, they’re just pieces of the puzzle. This is more than a technology problem. It’s also a people problem. Plan accordingly for surprises on both sides.
The IoT Technology Problem
You need to keep it simple. But not too simple. First, determine whether or not you can collect, process, and share the data you need. Second, learn whether or not this can be achieved with a reasonable level of resources or if you need to revisit the ROI calculation. Third, you want to understand the delta between your initial design and what it will take to procure the data at scale. That includes the velocity, variety, and volume expected in a production operation. Don’t proceed without understanding what it will take to operationalize your technology. You don’t need to build it yet, but if the amount of work required to move from prototype to production isn’t well understood, you haven’t proven anything.
Most Problems are People Problems
You can run small pilot tests to learn whether your technology changes will speed up the production line. Likewise, showing a customer remote monitoring data from your machines inside their facility is a quick and iterative process for collecting product feedback. Yet sadly, both are poor indicators of the overall value to be produced if the project is fully funded, developed, and deployed. Let me repeat that. Successful technical demonstrations and positive customer feedback are insufficient gates for moving ahead with system investment.
Often, controlled experiments don’t take into account the steps before and after the area of focus. Speeding up one part of the manufacturing chain can cause a total breakdown in another. Also, changes in technology require changes in human understanding. More importantly, they often require changes in human behavior. If service personnel don’t like the system, or if sales teams aren’t bought into the new model, nothing else matters. If the IT team isn’t comfortable with data security, you’ll never make it out of the lab. You need to get your internal teams on board with the project across your enterprise as part of the pilot. It’s not just about “things” – you’re optimizing a complete industrial system.
But wait, there’s more
Similarly, a customer saying they would use your new system is meaningless. The critical question to ask is “would you be willing to pay $x for this?” This is a crucial conversation that you must have for discovering what your customers actually value, not just what they like. It doesn’t matter if you’re wrong – unless you’ve already gone ahead and built a full solution that they don’t want. What’s the simplest thing you can quickly build and show them?
You’ve got to prove the value without requiring significant change. You’ve got to keep it simple. What’s the smallest combination of data collection, insight generation, and action triggers you can deliver for proving the value of your vision? A good rule of thumb is what can be developed and delivered in around 30 days. If you don’t have the right team, partners, or technology for creating enough value in a few weeks to get internal executives and customers on board, you don’t have the right team, partners, or technology to deliver a successful long-term solution.
Starting small while thinking big prepares you for scaling fast. In practical terms, this means your initial releases – designed to prove business value – are architected to grow quickly into production systems. Acorns are small, but they’ve got the oak in mind from the beginning. Don’t get your customers and executive teams excited with a demo that you don’t know how to scale. Once you open their eyes to the possibilities of optimizing their operations and generating new revenue, they’ll want to take advantage of the opportunity before their competitors do. If you need another 6-12-18 months to figure out how to build it “for real” then you’re going to lose the opportunity to others who are more prepared for growth.
The world is changing. Your competitors are on the rise. What new value could you provide your customers if you couldn’t just sell hardware? Where do you think the bottlenecks in your operations lie? Critically, what information would you need about your systems in order to improve them? Keep it simple. Find a problem that people understand, with measurable business value to solve. Build your roadmap, and take the first steps toward maintaining industry leadership in a connected world.
Originally this article was published here.
This article was written by Marc Phillips, the Director of Marketing at Bright Wolf, a leading IoT technology provider and system integrator helping Fortune 1000 companies design, develop, and deploy Enterprise IoT systems and connected product solutions.