Case study

Fiddlehead Delivers Predictive COVID-19 Market Insights to QSRs Using SafeGraph Patterns Data

Company

Fiddlehead is a data analytics platform offering insights into the consumer goods market. They work with leading logistics companies, including the top 40 global food and beverage manufacturers, to anticipate emerging customer, category, and competitor trends.

Headquarters

Moncton, New Brunswick

Industry

Consumer Goods
Contact Sales

The Problem: Predicting Consumer Demand During a Pandemic

How can data science companies offer timely and reliable consumer demand forecasting during a major health and economic crisis?

The Problem-Solver: Fiddlehead

Fiddlehead is a data analytics platform offering insights into the consumer goods market. They work with leading logistics companies, including the top 40 global food and beverage manufacturers, to anticipate emerging customer, category, and competitor trends. Fiddlehead uses deep learning to enable its customers to identify gaps in the market proactively and predict fluctuations in competitor pricing and promotions.

The Challenge: Assessing Pandemic-Driven Changes in Consumer Demand at Quick Service Restaurants (QSRs) 

The COVID-19 pandemic has essentially thrown supply chain forecasting models out the window. To get a better line of sight into rapidly changing market dynamics, Fiddlehead needed access to new data sources that could provide demand signals accurately representing pandemic-related shifts in consumer and competitor behavior. 

SafeGraph Patterns Data Fuels Reliable Forecasting Models 

SafeGraph Patterns foot traffic data allowed Fiddlehead to move away from outdated forecasting models—now made somewhat obsolete in the face of the COVID-19 pandemic—to create new and more accurate, what they call, ‘nowcasting’ predictive models. It provided them with an effective way to match foot traffic data against order data, in order to unlock valuable insights around stocking strategies and demand planning.

This made it possible to answer a critical question: Is a particular QSR location simply building inventory (vs. selling it) or is it actually addressing real-time customer demand?

‘Nowcasting’ quickly became a game-changer for them. They could now compare the relative performance of leading QSRs, to determine which locations were trending ahead of others, and also see where customer demand for certain QSR locations had bottomed out a full two weeks before stock delivery data would have normally been able to tell this story.

Now more than ever before, companies in the food and beverage industry need better ways to forecast demand, much less cope with disruptions caused by major “events” like the COVID-19 pandemic. Using SafeGraph Patterns data allowed Fiddlehead to take the lead in providing accurate, reliable, and trusted insights to help inform their customers’ commercial strategies. 

Additionally, Fiddlehead was able to leverage this enhanced analysis of consumer demand to help its customers with real-time stock delivery planning. They drew on SafeGraph Patterns data to identify when foot traffic (across various dayparts) was beginning to see a return to pre-COVID-19 levels, thereby helping QSR chains determine when they would need to re-stock on particular foods and, thus, better serve growing demand once again. 

Surveying foot traffic during the morning rush hours is a great example of this in action. Obviously, at the onset of the COVID-19 pandemic, foot traffic to many businesses, including QSRs, hit an all-time low. However, as social distancing measures have been relaxed around the country and people have begun to get back into their typical morning routines, Fiddlehead was able to use its ‘nowcasting’ model to help its customers gradually scale up their operations to meet increasing demand (i.e. stocking up more eggs and bacon to survive the breakfast rush).    

Insights like these have been particularly useful in helping QSRs place orders and manage shipment lead times in relation to ever-changing consumer demand. This has allowed them to take advantage of market gaps, plan future shipments, and avoid the risk (and loss) associated with overstocking perishable goods. Simply put, by weaving SafeGraph Patterns data into its forecasting models, Fiddlehead was able to provide its customers with a more precise roadmap for surviving and thriving in the face of the COVID-19 pandemic.

Since forecasting models are based on historical trends and data, they quickly became irrelevant when COVID-19 hit. We turned to SafeGraph Patterns foot traffic data to feed our models with an alternative data source representing demand signals that could radically improve downstream visibility and build greater trust with customers around our forecasting models.
Shawn Carver

CEO, Fiddlehead

Future Plans: Identifying Commuter-Focused Locations

Given the success of Fiddlehead’s ‘nowcasting’ models, powered by SafeGraph Patterns data,  the company is now exploring how to use this data to more accurately identify the QSR locations that primarily serve commuters. This will be useful as COVID-19-induced restrictions begin to ease and more people go back to work, potentially increasing foot traffic to certain QSR locations overnight. Being equipped with the right insights early on can help them avoid being caught off guard in the face of a sudden spike in demand.

Fiddlehead’s customers rely on them to provide the most accurate and timely market insights. To do this, Fiddlehead must have access to clean and current data sources at all times. For them, SafeGraph Patterns data was the answer—and it has proved invaluable in helping them create new insights while also building greater levels of trust and respect from their own customers.


Interested in SafeGraph's POI data?

View the data schema or contact sales for more information.