It’s always a tough time to own a restaurant.
In the best of times, a feeling of existential threat is commonplace in the restaurant industry. Between razor-thin margins, complex and vulnerable supply chains, inventories that are soon to become either “revenue” or “spoilage”, and the consumer’s shifting tastes and social habits, there is virtually no room for error.
These headaches feel like quaint nuisances from a bygone era. Months into the near-global shutdown brought about by COVID-19, with no certainty regarding timetables for the full relaxation of quarantine requirements, the potential fragility of social re-opening, and the willingness of customers to repopulate bars and restaurants in large numbers, there’s little distinction between surviving and winning. To “only” need to navigate the minefield of issues that threaten one’s business on daily basis would be a welcome respite for any restaurant owner.
There have been some feelings of relief, excitement, and optimism at being able to get back to business in some places across the country. But restaurants are coping not only with their standard battery of challenges, but doing so in the knowledge that many customers that do return are likely to be as conscious of cost as they are of physical proximity and, lest we forget, any setback in the fight against COVID-19 will likely result in the significant tightening of social distancing requirements, and possibly yet another order from lawmakers to shut their doors.
Against such a backdrop, it will be vital that restaurants have the best possible understanding of the factors affecting their business. The bigger and more complicated the supply chain, the greater the need for robust and reliable business intelligence. Fortunately, restaurants are hotbeds of behavioral and trend data. However, though they accumulate huge amounts of potentially predictive information regarding customers’ behaviors, converting this information into actionable intelligence isn’t always a straightforward process. It is thus paramount that restaurant operators focus not only on collating this wealth of data but managing, contextualizing, and analyzing it, to optimize forecasting.
For instance, behavioral research in the industry has shown over time that there is a significant correlation between the weather and customers’ purchasing habits. At a very basic level, the research has shown that as the weather heats up, customers tend to order fresher and healthier food items, such as salads. Conversely, when the weather is particularly cold or inclement, such during periods of heavy rain, customers tend to increase their purchases of “cozy” foods that are synonymous with staying home and “lounging” – this typically includes richer, heavier foods, such as burgers and pizza, as well as chips, fries, and other “junk foods”. Even in a vacuum, data such as this can be helpful. On a more granular level and analyzed algorithmically, such data can help a restaurant compile a profile of its customers’ habits and preferences.
Large food companies have made significant strides toward becoming data-driven businesses. Two prime examples of this Domino’s Pizza and McDonald’s, each of whom have made meaningful investments (in McDonald’s case, through the acquisition of analytics company Dynamic Yield) into technology integration and data analytics. This has allowed them to better understand the myriad tasks, internal and customer-facing, required to provide the service and quality that their customers demand. By gaining a better understanding of both their customers and their business processes, they’ve greatly enhanced their efficiency.
While these big players focus on optimization, at the other end of the spectrum, the primary concern for small businesses is forecasting. It’s here that the constant fight for survival is most evident. Rather than being able to focus on the nuances of each of their processes, smaller restaurants are far more concerned with the day-to-day. For these businesses the pressure to accurately predict both “what” and “how much” is immense.
The most common challenge on this front for smaller restaurants is one that has nothing to do with the impact of the pandemic, but that of fragmented legacy systems. Fragmented systems have long hindered restaurants in converting data gathered from operations, at various points throughout the business cycle, into accurate forecasting tools. For businesses with the wherewithal to move forward with such an upgrade, there are companies offering A-to-Z solutions using artificial intelligence to extract and aggregate fragmented data and apply algorithms to help restaurants control costs and increase efficiency. By combining disparate data into a coherent set, restaurants can identify their best- and worst-performing customer segments, target specific segments, identify trends, and adjust offerings based on real-time demand and product price and availability. This type of system enables business operators to proactively optimize the ordering of ingredients, food delivery, staff scheduling, reservations systems, and the management of vendors, inventory, and payables.
Beyond the simple question of investment, however, another potentially troublesome issue is that of data integrity. In many cases, longstanding restaurants (or chains) that have gathered data over their years in business become compelled to try and extract trends and intelligence. However, simply having data is not a solution, in and of itself. For starters, shifts in local dining trends, economic factors, and seasonal data (such as weather and local school/travel schedules) can result in data fairly quickly becoming outdated. Additionally, a majority of legacy businesses likely do not have any type of formal oversight of data quality in place, and thus run the risk of misidentifying trends using “unclean” data. Finally, there is a question of sampling.
What might look like a lot of data can often be statistically insignificant. Taken a step further, relying on a statistically insignificant, internally gathered set of data raises a restaurant’s risk of skewed results and the misidentification of trends. For smaller businesses whose internal data lacks in either volume or quality, turning to external data, whether gathered by social networks, financial institutions or dedicated platforms can prove more reliable. The ultimate example is OpenTable’s “Guest Center” analytics tool, which predicts, in real-time, approximate demand for various venue types, based on observed trends at restaurants of all types in a given region. By all accounts, however, the cost associated with “Guest Center” is prohibitive for a small restaurant. Fortunately or smaller businesses, similarly comprehensive solutions are provided by a pair of New York-based companies, Ingest.AI and Avero.
COVID-19 has taken a tragic toll on countless families and communities worldwide. The global response to the virus, combined with the impact of a prolonged quarantine threatens the very core of a U.S. restaurant industry that, as of a year ago, posted revenue of nearly $800 billion. For businesses without the resources required to weather an extended period with no revenue, operating more intelligently and efficiently, with an eye toward more nimble operations, cost controls, and optimized offerings for customers could be the key to not only weathering the current storm but also thriving in a post-COVID world.