What's in this article:
The world of the global supply chain has transformed drastically post-COVID-19. In supply chain terms, it was a global demand shock followed by a supply shock that began in China in February 2020. These successive demand and supply shock exposed worldwide vulnerabilities in manufacturing and supply chains. Temporary trade restrictions, labor shortages, essential products, medications shortfall, and convergence with the U.S.-China trade conflict increased economic nationalism during this period. As a result, global manufacturers now face more significant political and competitive pressures for localization, leading to cost increases. They are doubtful of reconsidering global lean supply chain strategies due to their fragility in uncertain times. All this has added immense pressure on the working capital of the companies. The combined effect has led to lower margins and difficult-to-forecast and tricky-to-plan supply chains. Supply chain resilience suddenly appears to be more important than agility and lean.
While supply chain challenges have increased manifold, customer expectations have increased even further. Customers want low pricing (particularly during an economic slowdown), faster deliveries, excellent quality, and consistent product availability as a standard now. Businesses with a supply chain that does not meet these expectations will find difficulties in winning customers.
The challenge for businesses will be strengthening their supply chains without sacrificing competitiveness. To face this problem, managers must first identify and recognize their vulnerabilities and then explore a variety of measures, some of which they should have done well before the pandemic.
It is no surprise that COVID-19 has significantly impacted the corporate world. This uncertainty is mainly because the last pandemic, the Spanish Flu, occurred more than a century ago. COVID-19 put numerous companies into unfamiliar territory. Remember how frequently "unprecedented" use was in 2020? This is the first time companies have encountered something like this in the digital era, and no quick-fix playbook exists.
Today, there is a need for a distinct supply chain strategy that considers not only previous trends but also other relevant variables of the current market environment. Therefore, established forecasting and supply chain planning tools and methods must effectively depict the market post-COVID during the ongoing epidemic. In addition, these forecasting and planning tools must be more flexible and capable; otherwise, they will be quickly irrelevant for future demand planning in uncertain situations.
By incorporating data analytics and machine learning into their forecasting, several companies can make more accurate projections than their competitors using conventional approaches. CFO.com revealed in 2018 that 74% of businesses that utilized sales planning analytics achieved better accurate forecasts.
However, the challenge for the post-COVID era is that while analytics might aid in producing more accurate forecasts, the projections will likely continue to be based on the same premise as pre-COVID. For instance, analytics data may track the number of encounters a potential consumer has with a product before purchase, but it may not notice that now the customer works from home and is more likely to order the product online than at a physical store.
Circumstances like the above may result in accurate overall demand forecasts but wasteful supply chain planning. For example, while the stores might have excess stock, warehouses are understocked and need help to fulfill online orders. In addition, they may need to account for future unexpected outlier events such as COVID-19 or Ukraine-Russia war. For this, they need to discard previous data that does not match the new normal in planning following such events.
Post-COVID, businesses will need to rely more on new-age AI/ML-based tools for forecasting. Utilizing organizational data with machine learning (ML) and artificial intelligence (AI) provides powerful means for enhancing demand forecasting. By using behavioral, intent, historical, and other data types (such as weather forecasts and social media trends) with AI, firms can assemble and evaluate current and expected future data to reveal more deep market insights. The best part is that there are no-code AI platforms available today.
On the other hand, Machine Learning enables forecasts to "understand" results and historical trends to deliver a refreshed forecast based on event discoveries or sudden demand shifts. In this manner, demand projections may be continuously refined and fine-tuned in response to market fluctuations with minimal further human intervention.
After two years of disruption, one of the most significant obstacles to planning is that supply chain managers must rely on more than historical data to create projections. Also, they must recognize the past two years and the customer behavior part of it.
For example, sellers began selling home gym equipment for double the standard retail price at the outset of the pandemic. The demand for home gym products increased so rapidly that manufacturers could hardly keep up.
But as the situation normalized and the outside world opened up, prices for home gym equipment began to go down. Now, companies are losing market share in this category, although demand has yet to return to pre-COVID levels.
If you're a supply chain planner for a company in this industry, you're likely aware that you can only rely on on-demand statistics from 2019-2020, as demand in 2021-22 was significantly greater. Demand today is different from before 2020, as customer behavior has changed. Therefore you cannot simply disregard this data.
Effective planning in the post-COVID age is possible, but companies may need a shift in planning tools, methodology, and data sources.
As the world becomes connected, consumer behaviors evolve and alter. To spot recent trends and behavior changes, utilize close-to-the-consumer data, such as point-of-sale (POS) information. It may be online queries or subscription patterns for online businesses. Using market-fresh data enables companies to respond quickly to fluctuating demand. Data that is further distant from the market, such as shipping data, can cause a delay in the company's reaction, costing revenue and market share.
Where monthly or weekly data sufficed before the pandemic, daily data will be more beneficial in providing rapid indicators of market changes. Even if the companies continue to plan monthly, detailed daily data enables them to produce forecasts closer to actuals. For online businesses, this allows them to quickly observe spikes or declines in inquiries and subscription purchases, which would be an early indicator of a possible demand trend change.
External data can provide information about consumer demand unavailable from internal data collected historically. Internal data frequently reveals what has occurred, yet more than historical data is needed to predict the future.
Consider addressing the following questions when analyzing external data:
How have places in regions where reopening policies were implemented earlier performed? Have they noticed an increase or decrease in client demand?
This knowledge enables you to incorporate the pre- and post-COVID shift in customer demand into predictions for those locations that are slower to reopen. For instance, if London opened first and there was a 20% decline in home fitness gear demand, the first step in planning for other regions would be to add this 20% decline into the projection model.
Similarly, in a changing and slowing economic scenario, questions like how will the gross domestic product (GDP) or inflation fluctuate? How does this affect the purchasing habits of your target market?
Using macroeconomic forecasts, your forecasting model may estimate how demand will vary in response to these anticipated patterns. Including GDP or inflation data in your projections will help identify trends and correlations between the demand for your product and the macroeconomic driver.
Or, what do top research organizations have to say about the demand in your market?
Research organizations frequently use numerous economic, demographic, and firmographic data to forecast demand. Adding this data to your forecast model can offset the demand swings caused by the epidemic or economic slowdowns, allowing you to predict future demand better.
Regularly incorporating new data and projections into your planning enables you to swiftly identify and respond to changes in behavior. This is only beneficial if the supply chain reacts swiftly to market developments. By moving faster than your competition, you acquire a significant edge.
If our home fitness company observed a decline in subscribers over a few weeks, they might modify the order size of their supply chain, reducing the amount of inventory in storage and the associated inventory carrying costs or the risk of obsolescence.
The last two years were challenging. Even industries that flourished during the pandemic had to change to meet the increased demand. The upcoming years can be better using AI/ML-based forecasting and planning tools. Utilizing additional data from internal and external sources will detect trends more quickly and effectively identify correlations between drivers and projections. Your data becomes plans when you can rapidly examine and react to this new knowledge. The capacity to make intelligent business decisions based on the strategy puts your goals into action. Using Ikigai, you can integrate vast volumes of data into your planning and forecasting process, systematically incorporating external data sources to produce precise predictions about future demand.
Ikigai is an operational BI platform that uses augmented actions to navigate toward company objectives. As the only commercially available product based on cutting-edge MIT research on AI and machine learning, Ikigai helps supply chain teams improve the speed and accuracy of their decisions in the VUCA world, thereby increasing the return on investment (ROI) of the business.
Ikigai combines data analytics, visualizations, and automation with its proprietary technologies, such as aiMatch and aiCasting time series forecasting, and translates them into more precise decision-making data. Let's connect for a quick chat.