Are your commercial industrial data projects stuck in Proof-of-Concept Hell?
Table of Contents
In a Prescient blog post from last year, we discussed overcoming challenges during edge data project adoption and getting your projects “unstuck” during their early stages. We explored how the path towards industrial data project success at the enterprise level traverses along three distinct phases: Visibility, Discovery & Transformation.
In the Visibility Phase, companies gather as much data as possible to form a hypothesis on how they can help improve their business
Once a hypothesis is formed, more data is gathered and analyzed during the Discovery Phase, often leading to revelations about how the business model can be altered to increase profits
Finally, if a company commits to the insights learned from their Discovery Phase, they can then go through the Transformation Phase; altering products, processes and messaging to significantly increase their market share and profitability
Companies who have successfully completed digital transformation/Industry 4.0 journeys have embraced this process, taken lots of small, quick steps and, perhaps most importantly, experimented. Furthermore, these companies didn’t just run one experiment - they ran LOTS of them. And they ran them during all phases of their edge data project adoption process, not just in the beginning. Some experiments, of course, worked better than others.
A key component of any experiment is testing and refining one's hypothesis. In the industrial data project world, this can mean doing a small pilot project or Proof-of-Concept (PoC). Since pilot projects are being worked on during all 3 phases of the digital transformation process, often with incremental results, companies frequently experience a variety of issues that can cause their projects to get stuck in Proof-of-Concept Hell and make them unable to move forward efficiently.
Over 65% of industrial data projects projects fail. How can companies avoid such failure and experience successful digital transformation?
How did a water dispenser company become a market leader using industrial data projects?
This company’s digital transformation was so successful that they ultimately became the market leader in enterprise water dispensing and grew their market share by more than 10x! Think of all the different PoC’s that they had to navigate along the way. Here's how they did it:
1. Visibility Phase: They used edge data projects for predictive maintenance.
In the Visibility Phase, the company installed water quality sensors in their dispensers to predict when a filter would need to be replaced. This probably started with a lab project. But what if the available sensor technology wasn’t providing accurate data? What if the sensor cost made the dispensers too expensive, reducing ROI to unacceptable levels? What if the data was inconclusive?
2. Discovery Phase: They focused on a new message
During the Discovery Phase, the company tried new marketing messaging focused on “Healthy Drinking Water”. This likely started with a small, regional campaign. But what if external factors lowered their success rate? How many versions of the message were tried before the best one was found?
3. Transformation Phase: They took a user-centric approach
During the Transformation Phase, the company added an LCD screen to their dispensers and a mobile app for their users. But what if the pilots for these features had taken too long and market windows were missed? What if, for example, the web application worked well on iOS but not on Android, cutting pilot feedback by 50%?
Do you have a project you want to transform? Schedule a demo with our experts today.
The 4 Most Common Types of Proof-of-Concept Issues
Clearly, the Water Dispenser Company didn't get stuck in Proof-of-Concept Hell and experiencing a few failures along the way didn't derail its path towards successful digital transformation. They're 4 common types of issues that tend to arise with Proof-of-Concept projects:
1. They Take Too Long
2. They Cost Too Much
3. Priorities & Requirements Change On The Fly
4. Goals Are Not Well Defined
1. Proof-of-Concept Takes Too Long
Like any lab experiment, a PoC requires observations, questions, hypotheses, methods, results, and analysis. Traditional project management methods requiring detailed, up-front specs have given way to leaner concepts, shortening project build-times dramatically. However, when you layer on the complexity needed to employ edge applications, cloud programming, dashboards, analytics, etc., industrial data projects can often require months of up-front development before they produce their first tangible results.
2. Proof-of-Concept Costs Too Much
With long development cycles come high development costs, often reducing the ROI on even the most successful industrial data projects to unacceptable levels right from the start. Outsourcing industrial data projects to more inexpensive, international engineering firms is a popular method of reducing project costs, often achieving savings of 30% or more. Setting up international development teams is another option.
3. Priorities & Requirements Change On The Fly
As discussed above, one of the hallmarks of industrial data project development is the ability to perform lots of experiments, changing targets and requirements on the fly based on the most recent outcomes. Traditional development methodologies and tools are very inflexible. When you tack on the costs involved with outsourcing development to a third party, the price of making agile changes rises exponentially due to time-zone differences and shifting partner-company priorities.
4. Proof-of-Concept Goals Are Not Well Defined
Overall, industrial data project goals are usually well defined but, sometimes, PoC projects are not. The non-linear and experimental nature of industrial data projects and digital transformation often makes it impossible to plan in advance for every possible outcome. And depending on results, the next step may always end up being a pivot. A flexible development system is a must to keep projects moving forward.
Move beyond Proof-of-Concept with Prescient Designer, and turn your prototypes into real-life applications. Start by scheduling a free live demo with us today.
How to Avoid Proof-of-Concept Hell with Our Low-Code Industrial Data Project Design Platform, Prescient Designer
Prescient Designer reduces industrial data project development time and cost, increases the ability to accelerate design changes, and enables the rapid iteration of solutions for optimized outcomes.
A new way to develop and deploy IoT systems that's 12x faster
Prescient Designer can speed up industrial data project development 12x faster, often allowing a single engineer to go from PoC definition to implementation in just a week or two. When combined with our pre-built solution templates to get users started, the process becomes even faster.
Using a Low-Code industrial data project development platform can also reduce development costs by 6x, or nearly 85%.
Prescient Designer also reduces development costs by 6x, or nearly 85%. What’s more, Prescient Designer allows program changes and redeployments to happen almost instantly.
Finally, Prescient Designer allows for rapid implementation, data gathering and analysis for rapid conclusions and iteration. The result is a quick redefinition and roll-out of a possible new Proof-of-Concept experiment in days, not weeks.
Conclusion
The emergence of Low-Code industrial data project development platforms like Prescient Designer can speed up development 12x, reduce costs 6x, and require 3x simpler support. Often, a single engineer can go from PoC definition to implementation in a week or two. The net result is rapid deployment of flexible industrial data projects that move forward, converge on a solution, and don’t get stuck in Proof-of-Concept Hell.
Schedule a demo with us and we'll show you how you can build 12x faster with Prescient Designer for your unique DataOps/industrial data project needs.