The proverbial “glass half full” perspective is one we often admire and even pursue. You might even be willing to place a bet if you know your odds are 50/50. But ‘half’ is not always good.
Let’s consider a high-stakes scenario – one that is frighteningly realistic for many in the grocery industry today – in which retailers only understand one half of the customer.
Let’s consider a more high-stakes scenario – one that is frighteningly realistic for many in the grocery industry today - in which retailers only understand one half of the customer.
This divide is caused by disconnected internal systems that lack the ability to support the nuances of grocery today. The severity of this challenge is magnified by the growing emphasis of fresh and ready-to-eat meal solutions. As consumers purchase more of these items, they also purchase fewer frozen, canned and packaged items. This change in demand impacts every aspect of retail. The planning, assortment, pricing, promotion and replenishment activities for a majority of categories across the store are influenced by the increased demand for fresh and prepared meal solutions.
Most demand forecasting systems cannot understand the significance of increased demand for fresh produce and how it affects frozen and canned vegetable categories, but the impact is significant and ripples across the entire retail value chain.
Your ability to profitably serve customers where you only have a partial understanding of their behaviors and preferences isn’t determined by how optimistic you are about the outcome. Success requires a complete view of consumer demand across fresh and center-store categories. So why don’t you have a complete view on customer demand? Read on.
Disparate systems create disconnects in consumer demand patterns across categories.
Increased demands in fresh are affecting other categories both in store and in consumers’ pantries.
A complete understanding of how demand affects the entire store is essential for effective demand forecasting.
“Viewing demand in silos - fresh or center-store - impedes the ability to profitably serve customers.”
Fresh and prepared foods are a competitive weapon in the market, but previous to this recent shift, consumer spend in fresh categories remained fairly consistent. Consequently, the impact of fresh on the rest of store was understood and predictable. For these reasons, it hasn’t been routine – and hardly a priority – for retailers to factor the impact of fresh into the demand forecast for the remainder of the store.
Now, however, with fresh categories driving nearly 49% of all dollar growth, grocers are accelerating their fresh positioning, purposefully deciding to leverage fresh categories for a competitive edge in the marketplace.
In other words, grocers perceive fresh as their winning ticket or their defense weapon against Amazon and other threats. And for good reason, as fresh is the fastest growing part of grocery retail today.
Sixty-seven percent of retailers report that sales in perimeter categories increased in 2018, and 80% expect their sales figures to rise again in 2019.
Fresh categories are becoming a competitive weapon for grocery retailers.
Increased focus on fresh means demand is no longer as predictable as it has been.
67% of retailers reported an increase in perimeter sales.
Supermarkets are promoting fresh more, varying their assortment of fresh items, and increasing in-store services, often offering ready-made food solutions like meal kits or grab-and-go items.
This is where things start to get complicated. Since it hasn’t been all that necessary in the past, retailers have neglected to connect the impact of ‘fresh’ dynamics to demand forecasting systems for the center store.
But naturally, when customers purchase more items from fresh categories, they will purchase less frozen, packaged and canned goods from the center store.
Consider a shopper’s weekly or monthly grocery budget, or their physical capacity to bring a combination of food and non-food items into their pantry or refrigerator. Those constraints stay pretty consistent, which means their inclination to buy from a particular category is directly linked to their purchase habits in another.
Retailers’ increased focus on fresh changes the way customers shop in and engage with the rest of the store, which makes it imperative that your demand forecasting systems harmoniously support the two. Operating in silos limits your understanding of shoppers as a whole - and your ability to profitably serve them.
Retailers have typically not put an emphasis on connecting demand forecasting for fresh and center store.
Siloed systems weaken retailers’ complete understanding of their customers’ behavior.
“Fresh changes the way customers engage with the rest of the store, making it imperative that your demand forecasting and surrounding systems harmoniously support the two.”
Achieving a holistic understanding of your customer improves the entire end-to-end supply chain, spanning categories and grocery departments.
Retailers often leverage best-of-breed solutions for the center store, but then manually maintain or use home-grown solutions to manage fresh. Or, companies invest in a separate vendor provider
to manage fresh, isolated from their center-store platform. The reality is, vendors that are great at both represent the exception rather than the rule.
Indeed, most fresh-specific technology is not built to understand non-food categories. And the systems in place for center-store planning don’t appreciate the nuances and intricacies of managing fresh items. Thus, organizations are often operating with disparate demand replenishment systems.
In fact, only 36% of supply chain professionals operate on a single supply chain platform. Even where this exists, demand is not complete.
You may have an understanding of one consumer’s habits in the center store and a separate system analyzing their fresh produce habits – but you have no way of connecting the dots to see how that one individual fluctuates between the two sides of the business.
Using separate systems to manage fresh and center-store categories creates gaps in demand understanding.
Today, most systems built to understand fresh cannot factor center-store categories and vice-versa.
Retailers must use a unified system in order to see the full picture.
“The rise of fresh presents considerable challenges for retailers’ current technology to provide an understanding of consumer demand.”
By only understanding half of the consumer, you only understand half of the items that are relevant and important to them, and therefore, only half of the demand. To grow profitably, your demand planning and forecasting systems must contemplate all the demand, across categories and departments.
There’s a simple test – ask within your organization, “Are the changes in demand for fresh categories included in the demand calculations for frozen categories?”
The answer to this question should not consider demand calculations by assumption or limited historical context but through actual computed demand halo and cannibalization calculations.
Even with best-of-breed center-store solutions in place, retailers are perplexed to find that the forecasts related to those categories are often incorrect; the predictions don’t line up to actual demand or sales results. That is because retailers only see half the picture.
Any influence that impacts demand will further complicate the process. Because demand can change so rapidly in fresh categories, retailers must account for and adjust to daily, intra-daily, and at times, hourly changes in demand, for each location, including warehouse and fulfillment centers.
Even the best forecasting solutions will fail if they cannot connect all categories, including fresh, into a unified view.
Rapid demands in fresh categories require the agility to respond quickly across all categories.
Even the most proficient center-store demand assessment methods aren’t accustomed to this level of granularity. And with this level of fluctuating demand, retailers can’t rely on historical data alone; linear analytics will not deliver accurate forecasts. They need something that does more than analyze past trends.
A recent study uncovers several issues that further complicate this half understanding:
Today, all of grocery is dynamic, but especially in fresh and prepared foods. There are many factors unique to fresh that customary grocery solutions aren’t built to understand.
Center-store tools aren’t innately equipped for data management associated with fresh requirements and factors associated with prepared foods and meal kits.
Sell-by dates, traceability for item recalls, limiting food waste, recipes, ingredient tracking, allergens, “rules of use” and substitution guidelines, nutritional information, country or region of origin... it’s absolutely critical to track these attributes, among many others, throughout the supply chain.
It’s an overwhelming amount of details, all at once, and demand for fresh fluctuates often, and by location.
The ability to correctly and rapidly process huge amounts of data and combine data sets is a requirement for any modern demand forecasting system.
Historical data is not enough - complete visibility is key to accurate forecasts.
The ability to correctly and rapidly process huge amounts of data and combine data sets is a requirement for any modern demand system
“Today, all of grocery is dynamic. There are many factors unique to fresh that legacy grocery solutions aren’t built to understand.”
To understand shopper demand patterns, you must understand all factors that impact demand, not just a portion of them. There may always be a degree of uncertainty when it comes to knowing the actual behaviors and tendencies of today’s consumer, but you can most certainly do better than half. With a unified view of customer activity,
made possible with the right technology, you can gain a full understanding of customer needs and motivations. Artificial intelligence can perceive all impacts to store-wide demand. In fact, you cannot achieve a full understanding of consumer demand without proven AI and machine learning.
Currently, 52% of retail supply chain executives say they spend too much time on data crunching – AI and machine learning solutions overcome this. Demand forecasting systems must also include machine learning, which drives continuous improvement of demand and forecast accuracy. AI can leverage massive sets of information from all sides of consumer behavior to help you understand the full picture. It allows you to track every movement and trend to shed light on who your customer is and what he or she wants out of their retailer relationships.
Retailers need technology that provides a unified view of consumer activity.
Most retailers report spending too much time manually manipulating data.
AI-enabled systems process and unify massive data sets, and machine learning provides constant improvement of data analysis.
Applied correctly, AI alleviates inconsistent inventory buys, overstocks (and the resulting markdowns), out- of-stocks, and margin erosion.
Seek out solution providers that have a proven track record in applied artificial intelligence. As a real-world example, a leading retailer had a 20% error rate in demand forecast for many years without meaningful improvements. This meant items were not selling as predicted 20% of the time.
An AI-based demand forecasting system with machine learning, however, reduced the error rate to 5% – that’s a 75% improvement.
The platform in this scenario is robust enough to carry the weight of both fresh and center store. And that level of improved forecasting accuracy, thanks to a holistic few of consumer demand, has a drastic impact on increasing margins and profits.
The correct application of AI can dramatically reduce issues like overstocks, out-of-stocks and margin erosion.
In a real-world application, one retailer reduced error rates by 75% (from 20% to 5%) using an AI-enabled solution.
employees across all continents
customers in 70 countries
of the top 25 grocery retailers globally