How Novel Prescriptive Benchmarking at the Local Level Changes the Restaurant Game

Press enter to search
Close search
Open Menu
Restaurant operators have faced three major challenges in using data to drive action. Prescription and actionability are paramount to driving successful outcomes.

How Novel Prescriptive Benchmarking at the Local Level Changes the Restaurant Game

By Jordan Thaeler cofounder of WhatsBusy - 09/09/2020

Ask any seasoned restaurant operator to show you their playbook and it goes a bit like this:

They begin by noting if store-level sales are declining. If sales are indeed down, they grasp desperately for market data to understand if it’s a macro problem or unique to their stores. Assuming they can even source market context (and most cannot), they start prioritizing actions that they think will steer the ship back on course.

Relevancy Market Data

This process has been fraught with challenges for too long, the least of which is the complete lack of relevant market data for the majority of operators (the key term is relevant, which we’ll clarify momentarily). If allowed to summarize, restaurant operators have faced three major challenges in using data to drive action:

First, the few benchmark tools that have been available lack granularity at the local level, neutering much of the benchmark’s value. For example, existing benchmarks report trends at the Direct Marketing Area (DMA) level across the United States, where there are 210 predefined DMAs. To demonstrate how absurdly coarse a DMA-level benchmark is, the entire state of Arizona is effectively a single DMA. That would imply that operator performance in Tuscon is the same as it is some-280 miles away in Flagstaff. This patently fails a basic logic test, which is why we said relevant market data is key, and that means data at a very local level. 

Yet even then operators should not simply compare themselves with restaurants in their local market because customers likewise do not simply lump all restaurants into a single bucket, and making decisions off such comparisons can be detrimental. This gets us to our next problem.

Second, restaurants have lacked consumer insights on their share of wallet, which if made available would inform them where else their current (and prospective) customers are dining. In homage to Clayton Christiensen, famed business consultant, customers hire restaurants to perform jobs. For some customers this is defined as a quick meal, for others it might be the experience, and there exist many nuances in between. In understanding the behavior of their customers, operators could more intelligently build a relevant, local benchmark that becomes much more valid for understanding performance and triggering appropriate actions at the right time.

Still, restaurant operations is not exactly a 9-to-5 job and there aren’t data scientists sitting idly by to undertake this work. This is why prescription and actionability are paramount to driving successful outcomes for operators, and this takes us to the final observed impediment. 

Lastly, existing benchmark providers have not invested sufficiently - if at all - in data science to provide operators with clear recommendations whose economic benefits are quantified ahead of time. With only so many hours in the day operators need to know what actions to take, what those actions are worth, and the return on investment of any taken action through the auditing of outcomes with data science. We understand why existing providers haven’t invested in such solutions: data scientists are expensive and restaurants are cheap. There, we said it. But restaurant operators need to be empowered with data science if they are to successfully compete with third party ordering and delivery companies, all of whom are actively investing in similar initiatives. 

The restaurant industry has long been hobbled by a lack of competitive data at a local level, a lack of consumer insights data, and the absence of data science to make insights actionable. If restaurants truly want to improve their decision making and drive financial returns, they must source tools that make them competitive in today’s environment. That means intelligent use of data should be the leading qualification in any vendor selection process.


About the Author

Jordan Thaeler cofounded WhatsBusy, a data science company in the restaurant vertical. WhatsBusy offers prescriptive operational and marketing recommendations on top of a benchmark seeded by POS data from 100,000 restaurants and consumer insights drawn from credit card spend at every merchant globally.