Think about the power of data in terms of the impact it can have.
You’ve probably seen statements like this for a while now. In our experience, this is an underestimate in financial terms alone, but it also misses the bigger picture. The right data enables entire new, innovative supply chain models. And, from there, business models.
There are many examples: a shift to “fast-follow” in fashion was not enabled by better trend forecasting, which had been the previous obsession of the industry. Instead, it came from creating a supply chain that was able to consume data at scale rapidly. Data which was used to identify existing products that were already selling well and flood the market with very similar products in very short timeframes.
This meant focusing on short lead times and emphasised predictability and agility in all aspects of the supply chain. This requires a huge amount of knowledge of each of the players in the supply chain and needed the kind of volume of signals that simply cannot be analysed by a human mind.
This is the first characteristic that we see in really powerful applications of data in the supply chain. Good data usage makes decisions about something that would be unthinkable for humans to do. Whether that’s because of the time required, the number of “brains” required, or the number of competing pressures to balance.
The second characteristic is equally important. It has to be coupled with the ability to act on those decisions within the relevant time constraints. If you set out with those two criteria in mind, you may be on to a winning formula that could fundamentally change your business. But unfortunately, that’s not enough.
Knowing the supply chain
You can think about this in terms of a shift in where knowledge is collected and codified, or where you’d go to make a decision. We typically find a very high proportion of knowledge in the boundary between people and process: a classic example of this is supply chain analysts and rules-based TMS systems.
In fashion, this could include data from stylists and trend analysts. This has a huge impact on decision speed and the time you continue making mistakes. To make a correct decision people and processes still require data. So you have to go and fetch it and analyse it.
Another aspect is equally important: coupling decision speed with the ability to act on those decisions quickly. Changes to processes require (if you’re lucky) re-programming and training. Meanwhile, changes to what people know and do require heavy change management programmes.
In this context data is slightly magical - a change in what you know through data is the same thing as a change in data - it’s instantaneous. And to act on it usually doesn’t require any change in activity by people, processes or software.
This means you can make decisions and act at much greater speeds and on much more challenging, high-value problems, in turn allowing you to rethink how your business can evolve its supply chain. And in Supply Chains this is more vital than ever: there’s more frequent disruption than we’ve ever seen in post-globalised supply chains and the size of those disruptions is increasing, and there are more and more cases of the supply chain becoming a competitive edge for business, so we all have to keep up.
Historically organisations have made strategic changes in their supply chains over multiple years whilst operational ones are happening daily or weekly. In the last couple of years, these timelines have shifted massively. You might be changing suppliers on a monthly time scale. You might rework your network on a 12-monthly basis. Operational decisions on logistics could now be in the hour-to-hour window. Tacit and embedded data is no longer fast enough to make good decisions quickly and this is unlikely to change.
What does it take to know your supply chain?
What is the right data that will enable this journey?
Typically, you must master five critical information sets: your demand over time and regionally. You need to understand your available to promise inventory and your suppliers and their ability and constraints to produce goods and you need to understand your logistics with particular exposure to transportation.
There are many many movements (and therefore many many dependencies) in the modern supply chain. Transport has complex system properties and sits beyond your organisation’s control or view which makes it even more critical to measure and is something quite often inadequately considered in data systems.
The final pillar is confidence levels and risks over time. This data set is a key ingredient to understanding how your supply chain will be affected when things don’t turn out as planned. It is important to have a number for demand forecasts and a confidence level of that demand. Where you have low confidence of forecast for a SKU you need a high confidence and capability to source goods from production and a high confidence to deliver those goods to where they are needed.
Codifying this knowledge in data rather than people is what makes it possible to act at the right speed to make meaningful adjustments in meaningful time frames.
Managing the transformation
Getting all of this right sounds like a monster transformation project.
But there is a right way to bring data to bear on your supply chain that makes this much lower risk and if done right has much faster time to value.
Typically, you’d need to master three critical information sets: accurately forecasted demand and risk at lead times, production capacity and capability, and both inflow and outflow risk.
Supply is relative to Demand. Demand is composed of many channels and variables of varying complexity to forecast. It may in fact be easier to forecast and act on the uncertainty associated and reduce reaction times than “forecast better” at the same lead time. Supply is also composed of production and logistics. Logistics is particularly exposed to transport, because of multiple movements and dependencies in the global supply chain. In the modern supply chain, transport has complex system properties and most often sits beyond your organisation’s control or view
Crossing the gap from data storage to data-driven
But to cross the chasm of failed data projects that deliver known value it's not enough to just know the supply chain and have the right data available, you have to be able to do several things together. Make data the centrepiece of your decision making: you can’t be data-driven if it’s window dressing. This means working out the collection of data you need in one place to solve a defined question. But that means you need to know which problems are valuable. This means you need a broad data set too! So you need to look for this in your data:
- The decision capacity and capability to match the time horizon for each decision
- The tools to decide (and importantly also) execute
- Don't go broad, go narrow and build on it incrementally (if you try to go for a big bang approach or add more headcount your solutions won’t scale and will be highly costly at best and at worst you’ll join the graveyard of software projects that couldn’t get off the line)
In each incremental case, deploy the deep capability to acquire data, process and use it in the right timeframe.
Finally, know and communicate the commercial value of each iteration. Validate this value quickly and forecast the value of the next step. This communication will drive company-wide adoption and build confidence in the transformation you are undergoing. The end goal is sustained transformative value that is truly data-driven.
How do you do this more practically?
This is an example of the engagement process a client has with our software.
It starts with building a data layer. Transportation is not only one of the highest sources of volatility in a global supply chain but it can also provide a very rich data set to explain a supply chain and measure its performance. We start by capturing invoice data from first, middle and last mile transportation. This gives us a view baseline view of how the supply chain is connected and performing.
We use the invoice data to:
- Create a data layer - which doesn’t require IT input
- Measure current performance vs target metrics and hold suppliers to account
- Finding room through savings to fund your current and future projects
- Use this data to build a digital twin of your network
- Run simulations to figure out what is possible and explore your problem and solution space
- Use output Business case to create a transformation roadmap prioritised to create regular funding and keep your project cashflow positive
- Move to roadmap execution using a single software platform to minimise training, integrations, and change management
Evaluate results with the data layer
We started 7bridges by looking at supply chains with a data-first lens. Our experience as a founding team was that supply chains were highly non-data-driven and full of inefficiency and low-performance systems. At the time of founding 7bridges, the power of data around the world was clear in many industries, but not supply chains.
At the heart of it, however, supply chain’s problems are mathematics-based. If you can leverage data properly and turn it into action, the value is immense.
Indeed, if you ask anyone with a supply chain qualification you’ll most likely hear the 7bridges of Konigsberg problem after which our company is named. This is a problem about connected networks and how to navigate them which are the problems we’re solving at 7bridges. And we know that smarter, connected supply chains benefit everyone.
Let’s talk about solving your supply chain problems with data– get in touch today.