Today’s Supply Chain is a data-intensive operation. All aspects of the Supply Chain function leave behind a digital footprint in the form of orders, movements, inventory and fulfillments. All organizations have a wide variety of data sources including Forecasting, Shop Floor, Relationships with Suppliers and All types of Documentation including spreadsheets. Accessing all this information is not sufficient to make effective Supply Chain decisions. It needs to be connected and to be understood.
Just having more data does not automatically translate into better supply chain decisions for companies. A lot of the information collected by different parts of a supply chain is difficult to manage in its raw form. Connecting data points, making sense of them and using the supply chain information in time to affect change is a challenging task that few companies have mastered at present. Supply chain analytics resolves these issues by creating a model of a supply chain in order to analyze information collected from throughout the chain to aid better decision-making.
When a company does not want to, or cannot, develop required supply chain analytics capabilities in-house, it can consider outsourcing these activities to a specialized supplier. If, however, the chosen supplier fails to understand operational and commercial specifics of data that is analyzed by this supplier, then expected value from analytics will not materialize.analytics.
Kepler Advanced Analytics is different from other potential Supply Chain Analytics Outsourcing providers.
Supply Chain Analytics Is About More Than Dashboards
While a dashboard can show increased inventory, higher transportation costs, or a decrease in supplier delivery performance for a supply chain manager, the key questions are what causes the issues and what can be done about them?
Increasing inventory levels – is demand decreasing, are purchasing quantities too high, or is there a change in lead times for certain products? Rising transportation costs – are they due to higher freight rates, less frequent shipments, sub-optimal routes, lower carrier performance, or even a change in the overall distribution network? Suppliers not meeting expectations – does this affect production or customer service, or is it an overall supply risk?
Understanding why things happen in the first place is the starting point for any successful supply chain manager. And in order to answer those kinds of questions, the simple production of reports in the form of so-called dashboards is not enough.
As said before, analytics should not generate more reports for the supply chain manager than he already has to deal with. It’s important to connect data with the business context of the analyst and make decisions that are based on experience of the processes of the supply chain. In other words, we want to go from description of facts to decisions that will lead to further actions.
Analytics supporting the supply chain must enable the supply chain team to make better decisions. Thus the purpose of supply chain data analytics is not to produce more reports than the organization already produces but rather to understand the reasons for supply chain performance metrics and to forecast future events that may impact the organization’s supply chain.
Why Are Companies Looking at Supply Chain Analytics Outsourcing?
For companies that want to go down the internal route, building a supply chain analytics capability from scratch requires skills in supply chain processes, data engineering, analytics engineering, data visualization, statistical analysis, demand forecasting, and advanced analytics techniques to analyze complex data from various sources in the supply chain. In addition, a solid data environment needs to be in place to store and manage the data that is required to generate meaningful insights from.
Most companies will at some stage require a combination of skills within supply chain analytics, which can often be too hard to develop in-house while the organization is still developing its analytics maturity. Supply chain analytics outsourcing can provide the required specialized analytics capability on an as needed basis.
Contracting supply chain analytics services helps organizations to meet specific analytical needs and keep their own analytics function down to size. At first sight, the required analytics solutions seem to be fairly standard: analysis of inventory data, of supplier data, of logistics data, of demand data, of supply chain reporting data.
Even when data is stored in various systems, good analytics outsourcing allows for the creation of a powerful analytics environment for supply chain analytics by compiling all necessary expertise for that specific task – data engineering, analytics, supply chain.
Strengthen your organization’s use of data and decision making.
Supply Chain Data Requires Supply Chain Understanding
One of the most important factors in the partner selection process for supply chain data analytics is the business understanding of the supplier’s processes.
At its core, a data analytics team is not necessarily going to understand the intricacies of the Supply Chain processes, therefore they can’t possibly begin to build out meaningful Analytics around a company’s data, until they understand the business process behind that data.
It is critical to recognize that supplier performance analytics (e.g. for on-time delivery) contains hidden nuances within each category of performance. For example, a supplier rated as performing poorly in terms of on-time delivery may actually be a supplier of strategic parts or be contracted under specific terms for lead time to that production facility.
Similarly, an unusually high inventory level does not have to be a problem. Many factors can influence inventory levels, for example safety stock, long lead times, seasonal peaks, Service Level Agreements and more. Analytics can support the Supply Chain Manager in understanding these factors and building adequate planning processes.
Analyzing supply chain data requires more than just processing information in order to generate reports and graphs. Supply chain management analytics must also be well versed in terms of the various processes in place for procurement, operations, inventory management, logistics, planning, and working with suppliers.
Why Kepler Advanced Analytics?
Kepler Advanced Analytics is the data and AI practice of Kepler Consulting Group. It brings the best Supply Chain data analytics experts to work with the company’s vast expertise in Supply Chain, Procurement and Operations.
Many of the Supply Chain Analytics Projects we work on sit at the intersection of technology and business. So while the data may reside in ERP systems, procurement platforms, warehouse management systems, etc. the questions being asked are operational and commercial in nature.
Is the data from an ERP system, procurement platform, warehouse management system or even a simple spreadsheet? Who are the users? Are they Procurement leaders wanting to understand average spend and average purchasing frequency by supplier? Are they Supply Chain leaders analyzing average inventory and service levels across the organization? Operations teams trying to determine root cause of a varation in their processes or logistics teams wanting to determine primary drivers of transportation spend?
This connects what they are trying to analyze with the actual data and enables them to arrive at meaningful insights that can be used to make operational and strategic decisions in their respective functions.
The analytics are developed for the business problem at hand using the relevant data to produce meaningful insights for the supply chain to make decisions from.
Combining Supply Chain Expertise With Data and AI
As mentioned before, Supply Chain Data Analytics has many layers of work in it. The first step is to set up a data consolidation platform to combine all relevant information from all different systems, such as ERP systems, procurement platforms, warehouse management systems, transportation systems, supplier databases and spreadsheets. After that, we need to set up consistent definitions for all different variables involved.
Once you have combined all your relevant data into one data source, you can start working on the analytics engineering. This stage includes setting up appropriate data models in order to do appropriate analytics for your specific questions in supply chain management and enable you to do repeatable analysis and create reports.
From the basic reporting to dashboards, supply chain analytics goes to very advanced forecasting, predictive modeling, and even to a full optimization for a particular process or function of the supply chain.
It is also important that the technology selected follows the business need of the organization. In some instances, a simple and reliable supply chain analytics dashboard will provide sufficient value to the organization when it brings together all of the information previously scattered across innumerable spreadsheets. In other cases, the organization may require predictive or even prescriptive analytics.
It is important to select appropriate forms of analytics for supply chain challenges.
Where Supply Chain Data Analytics Can Create Value
There are many areas within the Supply Chain where data analytics can add value, across many different functions within an organization and within different industries. Prioritisation is key, depending on the specific organization’s operating model, industry and data maturity.
Inventory Analytics
Inventory analytics helps companies analyze their stock levels and movements, and also indicates situations such as excess stock or slow-moving products and tries to identify the root cause of excessive inventory and interdependencies with demand.
Supplier analytics
Supplier analytics involves amassing supplier performance, spend, delivery, quality and other data that enables the procurement and supply chain teams to gain a better understanding of their suppliers, including areas of supplier risk or where supplier performance is decreasing.
Logistics analytics
This type of supply chain analytics studies transportation activity, delivery performance, and shipment and costs. In the context of our keyword research, this is obviously highly relevant as there is significant demand for information on the subject of logistics analytics and data analytics in logistics.
Demand and forecasting analytics
Historical demand patterns, demand variability, relationships between demand, inventory and supply. If there is sufficient data available to build accurate models, then predictive analytics to forecast future requirements.
Warehouse analytics
Warehouse analytics supports the analysis of a warehouse’s performance with respect to the stocks that are held there, the inbound and outbound movements of inventory, the capacity to receive and ship orders and the overall process of order fulfillment in the supply chain. It is critical that warehouse information is combined with inventory information, order information and logistics information to provide true understanding of how the warehouse functions within the greater supply chain.
Supply chain risk analytics
Identifies patterns and risks within supplier dependencies and possible disruptions. Though no system can eliminate all risks, it does help to highlight areas of the supply chain where special attention is needed.
From Fragmented Data to Supply Chain Insights
One of the key challenges to performing Supply Chain Analytics is that the data that is required typically resides in multiple fragmented locations.
A company’s supplier data may reside in a separate database system than the information regarding their purchasing activities. Furthermore, the inventory and order information is typically stored within an organization’s ERP system while the corresponding transportation and operational data in the logistics department is maintained in standalone systems. In addition to these systems, countless spreadsheets full of operational data are distributed throughout various departments within an organization.
Much of the above work can need to be done before any form of complex analysis can take place.
To enable the necessary analytics, data engineering, analytics engineering and knowledge of the supply chain processes need to be combined. The structured data sources must be first of all consolidated, cleansed and then re-structured to enable the necessary analyses. By defining the right metrics and the necessary links between the different data sets, the resulting dashboards and corresponding analytics models can then be set up to support the supply chain decision makers in the best possible way.
This data can then be more easily used to create useful supply chain dashboards and analytics.
Kepler Advanced Analytics views the transition from data to supply chain analytics in a step-wise manner. Clearly, a simple deliverable (e.g. a dashboard) is not sufficient, as the underlying information needs to be of sufficient value to support the people that make supply chain decisions.
Data sources → Data integration → Data quality → Analytics engineering → Analysis → Visualization → Business insight
Thus, the end goal of your supply chain information does not have to be the supply chain information dashboard itself. Instead, the supply chain information must be useful to people responsible for making supply chain decisions.
A Global Supply Chain Analytics Delivery Model
As a company operates globally, having analytics performed for a client in different parts of the globe with different cultures would be vital to Kepler’s success as well as aid in their ability to understand their client’s specific needs in regards to their supply chain and more. Having 4 offices throughout 3 continents allows Kepler to support large global supply chains that have data and decision-making spread across countries, regions and business units.
The other major consideration is the partner’s global geography to serve their customers, as many large, multi-regional multinationals have their supply chain teams in many parts of the globe. Kepler’s global footprint includes 4 offices on 3 continents in the following locations: Detroit, Michigan (serving United States, Canada and Central America); Paris, France (serving Europe); Shanghai, China (serving China and Asia); and Chennai, India (serving the rest of the globe from this location).
For companies with multiple countries or business units, analytics requirements may first start off at one location but end up requiring support from analytics models at multiple markets.
A global delivery model can support an analytics outsourcing engagement that has requirements spanning multiple countries or even markets.
A Practical Approach to Supply Chain Analytics Outsourcing
Large analytics transformations are rarely the starting point for organizations who have a number of different potential analytics projects underway. Trying to tackle too many analytics opportunities at once can make the initiative overly complex.
Rather than looking at large analytics transformation projects as potential candidates for supply chain analytics outsourcing, more sensible approaches start with a specific business challenge.
This initial project is best framed around a single business problem such as an analysis of inventory levels and the causes for continued increases. For the supplier, one could analyze the required data to better manage suppliers. For logistics one may look to analytics to reduce the cost of transportation and for planning organizations one could develop a supply chain performance measurement report that replaces manual reporting of current performance.
First, build out a working model (often similar to a first project) that has a solid data foundation, and a good understanding of the various supply chain challenges. From there, they can start to add in other necessary pieces of the complex puzzle.
So while the short-term aim of outsourcing is to find someone to do the reporting for you, in the long-term there is so much more that can be done with a good analytics supplier.
Choosing the Right Supply Chain Analytics Partner
Note that choosing the right supply chain analytics partner for outsourcing is more than choosing a supplier for producing dashboards.
Ultimately, the organization is looking for a partner that can understand the problems of the supply chain and relate to the required analysis on the enterprise data often scattered over various systems (e.g. billing systems, ERP systems etc.). They should have enough expertise in data preparation as well as analytics to perform Descriptive, Predictive and even Prescriptive analysis, depending on the problem at hand, and grow with the organization and its required analytics evolution.
Remember that supply chain analytics projects are not typically isolated events and require evolution as the data, business environment and even supply chain itself changes.
Outsourcing its supply chain analytics to a partner could go in different directions for a Supply Chain Organization; having it deliver a one-off report is not the only option. The more important issue for long-term success is that the supplier can handle evolution of required analytics over time.
Building a More Data-Driven Supply Chain
There is considerable information about a company’s supply chain already in place and therefore the main challenge is to make this information accessible and to ensure that it is consistent and of use.
Supply Chain Analytics links together data from across all the different functions in your supply chain including Procurement, Inventory, Suppliers, Logistics, Warehouses, Planning and Operations functions.
Supply Chain Data Analytics looks at the underlying causes of performance and provides insights whilst Advanced Analytics can be used for Forecasting, Risk Analysis, Scenario Planning and even Optimization.
There are many companies that are depending on supply chain analytics to improve their supply chain but cannot build up all the necessary capabilities internally. Thus, they are considering supply chain analytics outsourcing.
At Kepler Advanced Analytics, we have long experience in supply chain and in procurement. We combine that expertise with world-class skills in data analytics, analytics engineering and, above all, AI.
We operate from four centers: Detroit (USA), Paris (France), Shanghai (China) and Chennai (India).
Kepler Advanced Analytics helps to get value out of the mass of data that is strewn throughout the modern supply chain.
One does not have to start with a massive transformation of all analytics activities within your Supply Chain organization.
The foundation for a great analytics capability is one business problem, one data challenge and one decision that you want to make better.
From there you can go and scale up to the whole supply chain.