Retail planners need an agile approach to see them through a pandemic, with artificial intelligence and machine learning techniques among the tools to help make the best decisions at every stage from initial response to recovery.
The changes caused by the arrival of Covid-19 in the UK have been dramatic, as focus is shifted heavily to public health and the safety of our communities. Barely any industry has been left unaffected, but for retailers the upheaval has been all-consuming.
Whilst the consumer packaged goods (CPG) industry struggled to react quickly enough to dramatic increases in demand as customers stockpiled at the start of the pandemic, fashion retailers face an entirely different challenge. Shops have been closed for months and, for many, demand will have taken an unprecedented fall. With inventory piling up, a whole season’s worth of stock is out of date. As shops now get set to reopen in a new world of social distancing and unknown demand, merchandise planners are faced with some of the trickiest decisions in modern commerce.
However, it’s possible to predict the most likely outcomes of retail planning actions. Advanced forecasting capabilities can leverage artificial intelligence and machine learning techniques to capture and model the impact of retail decisions. Understanding the four distinct phases of the pandemic is the first step to managing demand; these stages are preliminary, outbreak, stabilisation and recovery. The phases highlight two vital actions for retailers: respond and recover.
Each phase has a different impact on demand, meaning retail planners need an agile approach to see them through the pandemic. There are also lessons to be learned about how to stay alert and ready for future crises. Below, we discuss each phase, the impact it has upon retail and how planners can deploy advanced forecasting to make the best decisions with perfect timing.
Figure 1. Illustration of Covid-19 phases and retailer demand across brick & mortar and e-commerce channels; overlaid with actual Covid-19 daily cases data from China.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataRespond: The preliminary phase
A pandemic begins with the preliminary phase. In this phase, it’s business as usual, with no government mandates and fairly typical consumer behaviours. Hypermarket retailers may see a slight increase in sales as a small percentage of society will begin to ‘stock up’ on core essential items. Most retailers will not see an impact to demand.
With business as usual, forecasting utilises trends, seasonality and cyclical patterns as well as causal factors such as promotions and in-store events to drive a statistically based merchandising process. Advanced machine learning techniques, along with location and merchandise attributes, intelligently develop sales forecasts for new items or items with little to no historical demand.
Merchandise planners can utilise the statistical forecast to plan overall sales within the merchandise financial planning process. Buyers can leverage this demand signal to plan core assortments as well as fashion assortments. The forecast can also automatically integrate into current replenishment and allocation solutions to intelligently drive inventory management decisions. Each week, as sales actualise and new data is available, a forecasting engine will re-forecast demand, taking into account recent trends. As the preliminary phase moves into the next phase, this becomes an important capability for driving better insights and quick responses.
Respond: The outbreak
The outbreak phase quickly follows. Consumer demand begins to depress, retail locations begin to close due to lockdown orders, and eventually, this leads to company-wide store closures.
As these events occur, retailers are challenged with understanding how to plan for this rapid decline in demand. One of the best ways to do this is by leveraging an analytical forecast modifier. This approach intelligently modifies a current forecast. This current forecast could be a statistical forecast, an open source forecast, a naïve forecast or even a working plan value from Excel. In essence, it has the ability to intervene and adjust everything from merchandise plans to replenishment decisions in unprecedented situations.
This analytically derived adjustment utilises external data, such as Covid-19 statistics from governments and public sources such as the World Health Organization (WHO) and Centres for Disease Control (CDC). There are also programmes available on GitHub that provide data and scripts that automate the refresh of this data.
Recover: Stabilisation
As the number of new disease cases begins to plateau, the pandemic moves into stabilisation. This phase is characterised by governments gradually removing restrictions across regions. The challenge for retailers becomes how to analytically capture the impacts of the pandemic on past demand, while building realistic plans going forward. As demand starts stabilising, that’s when event modelling across the different phases of the pandemic, and incorporating causal variables into forecasts, are critical to ensuring valuable demand forecasts in the future.
Retailers can leverage past event data from natural disasters such as hurricanes, along with demand data from markets that were hit early in the pandemic, such as China, as proxies for the expected pattern of demand modelled into an event. From there, retailers can run what-if scenarios on the next phases of the pandemic – including the event duration and recurrence in the future – to understand best- and worst-case outcomes proactively.
Improved forecasting during the stabilisation phase also includes incorporating both internal and external causal variables. Internal data includes sales and inventory history to capture demand shifts across time, pantry loading effects, and long-term product mix changes. Product and location attributes can be leveraged to understand changing consumer preferences both during and after the event. Understanding website traffic along with store closure and re-open dates regionally is important for capturing the shifts in demand across channels. Retailers can incorporate external variables as well, including historical Covid-19 statistics, future Covid-19 projections, and government restriction dates.
During this important phase, machine learning algorithms assess demand shifts at the most granular levels, and automatically learn, adapt, and improve over time as new demand data is available. This provides superior forecasting results, leading to improved inventory productivity and consumer-centric decisions across both merchandising and supply chain.
The stabilisation phase is still categorised by uncertainty as to what the future holds. Therefore, it’s imperative for retailers to have an agile, iterative forecasting process for capturing new data as the situation changes from day to day. Great forecasting systems automate the forecasts and quickly react as new information comes in, creating an immediate impact on downstream decision-making. This positions retailers on a path to recovery, and ultimately, to long term growth.
Recover: The final phase
In the recovery phase, while social distancing rules and regulations may still be in place, companies now have the confidence to plan their business without major constraints. Accurate machine learning statistical forecasts to drive business decisions are essential. As models were improved in the stabilisation phase with additional causal variables and events, reviewing and re-tuning models to improve forecast accuracy is now the core objective.
Incorporating new information into machine learning models allows users to understand what variables will be useful in explaining recent demand shifts due to the Covid-19 virus. Evaluating forecast models may involve understanding the significance of casual variables such as Covid-19 cases, unemployment data, store traffic, and other useful information that was introduced in the stabilisation phase. These models will need to be constantly re-tuned as demand patterns remain unsettled.
Reimagine: Prepare for the future
Forecasting is hugely important to retailers at every stage of a pandemic. Not only does it pave the way for a smooth re-opening, and a potential surge in sales, but it helps brands manage fluctuations in demand and customer behaviour throughout the disruption to normal service. Bringing together both internal and external data, models produce valuable insights into demand across products, locations and channels. In turn, this drives excellence in merchandising, financial planning, replenishment strategies and much more. Though lockdown has caused significant changes for many companies, with advanced analytics it is still possible to optimise apparel sourcing and supply chain operations, to recover quickly from the pandemic and go on to future prosperity and success.
