How digital twin can change the face of retail (It’s easier than you think)
73% of data generated by enterprises goes unused in decision-making.
That's a LOT of wasted data.
Could the best way to leverage retail operations data be a virtual counterpart to a physical process: a digital twin?
Digital twin is not just a buzzword; it holds a lot of power and opportunity for retail organizations to improve their operations and get real-time information about the customer in ways that other data analytic strategies can't.
Our resident digital twin experts, Mason McClelland, and David Holland, recently discussed the purpose of digital twin, why it's an option many retail companies are seeking out, and how to get started if it's the clear next step for your company.
Check out conversation highlights below:
Q: What are the greatest challenges that companies face with their data?
A: Not only are companies not getting the most out of the data they do have, but they're also not bringing in external data to make decisions. That means that they're not able to do test and learn experiments in order to start generating insights — they're just not using the data in their hands.
"Digital twin helps us to use more of the data we are generating to serve our customers and operations better."
- Mason McClelland, Director of Strategy
Q: What is a digital twin?
A: At its core, a digital twin is a copy of a physical process or product that gives us the ability to really see what's going on while also having the ability to experiment with new tactics and new approaches.
A great example I like to use is a reference for any Drive to Survive fans out there. F1 cars were using digital twin fairly early on in its inception. When you consider the massive size and very expensive price tag on race cars, it’s important to be able to quickly test design changes and how new parts twould affect the aerodynamics and performance of the car.
By creating a digital representation, they were able to simulate the effects of changing the wing of the rear of the car to see how that would affect aerodynamics and how it would speed up or slow down on certain tracks.
This is a highly growing market. In 2020, digital twin was estimated to be about a six and a half billion dollar market; projections are showing that number growing to 35 and a half billion by 2026.
Q: What does digital twin mean for the retail space?
A: The way I visualize it in my head is this: you're getting a map of your store or footprint of stores and the ability to understand what's going on in all of these locations. Whether that's the movement of goods originating all the way up in the supply chain and going through warehouses, how the goods are moving on the product floor, how customers are making visits through the store, or how they are interacting with products.
On the operational side you can also ask, where is my staff located? How am I positioning them to best serve my customer? These questions are answered with a digital twin.
“We’re talking about a perspective shift, the tools that we have to look at historical data. Digital Twin is a way to turn that around and start looking forward.”
- David Holland, Director of Technology
Q: Why is digital twin something to consider?
A: The platforms that are becoming available to really unify data sources are enabling scenarios that just weren't possible before. A lot of tools like RFID, robotics, IoT sensors, embedded analytics, and APIs are not only very powerful, but they're also becoming more accessible to the existing teams that you have while using existing skill sets. It's not inventing new technology, but instead, finding the right use case for all of these different technologies that we've already put together.
Q: How do I actually implement this in an organization?
A: There are data sources that exist that have to be accounted for in the systems that are used. You need a place to store all of that data so that you can work with it and use it to start building models. The next step is where the digital twin model itself comes into play. It's going to sit on top of that repository and make use of that data to inform the ML/AI processes that are running to help you to make those predictions. From there, it can be taken into the various forecasting tools, analytical tools, and other systems.
Q: What is the best strategy to use in approaching digital twin?
A: We have a saying at RevUnit, “Build small, learn fast, and iterate often”. Often when we think of big technology efforts, we think that we have to do everything all at once. What I would encourage you to think about is, what is the smallest problem you can solve? Then, by growing that use case over time, you will build acceleration and adoption within your enterprise. Ultimately, these solutions can be as big or as small as you want them to be.
Want to continue this digital twin conversation with a data expert?
Ready to ask Mason and David more detailed questions about your organization to see if digital twin is right for you? Let’s chat.
In the meantime, check out our article about Improving Data Quality for Better Decisions Making.