Watching weather’s impact on consumer behavior since COVID-19

New research from AccuWeather and Foursquare

Jacek Dylag Unsplash

Changes in weather–from rain storms to heat waves—can significantly impact people’s behavior. However, marketers no longer need to make assumptions about how the weather may or may not affect business results. Accurate first-party datasets around weather and foot traffic can verify weather’s true impact, proving out hypotheses and uncovering unexpected trends.

“Weather can be a powerful factor in driving foot traffic to businesses,” said AccuWeather Senior Vice President, Advertising Sales Bill McGarry. “For example, by having the most accurate weather forecast, retailers and QSRs will be in a better position to calculate how and when to take best advantage of warmer, sunny weather that naturally lends itself to more foot traffic, and when they must adjust to changing consumer habits as the weather turns colder, especially in the era of COVID-19.”

To quantify the impact of weather on visitation to various places, we’ve combined AccuWeather’s weather data with Foursquare’s proprietary foot traffic panel of millions of Americans to analyze real world behaviors. We examined data from both 2019 and 2020 so far, to uncover how the role of weather has changed since the COVID-19 pandemic. We took a closer look at behavioral patterns in three cities – New York, Atlanta and Denver – to learn how weather dynamics influence different geographies. For example, since New York is such a walkable city, we might expect weather to affect foot traffic more than other cities. People in Denver tend to favor outdoor activities like hiking, and so we might expect a stronger correlation between the weather and visits to trails. A sprawling metropolitan area, we might expect Atlanta to fall somewhere in between.

The Takeaway?

As winter approaches and the weather cools, savvy marketers will leverage these insights to predict behavioral shifts and adapt their operations accordingly, from inventory management to open hours to delivery options. With the power of these combined datasets, brands can better understand how both temperature and precipitation influence peoples’ movements—and ultimately, business outcomes.

Let’s take a closer look at the findings:

New York State Of Mind
In order to understand how temperature affects consumer behavior in New York – and how that has changed since the COVID-19 pandemic – we analyzed correlations between foot traffic and the AccuWeather RealFeel® Temperature, an exclusive metric using multiple factors including the air temperature, humidity, cloud cover, sun intensity, and wind to explain how hot it feels outside.

So where do New Yorkers visit more when it feels warmer outside? Location data shows a positive correlation between temperature and foot traffic to restaurants and bars, meaning people in New York are more apt to dine and drink out as the weather warms. A 10° F increase in the AccuWeather RealFeel Temperature typically resulted in a 5% increase in foot traffic to restaurants and a 4% increase in foot traffic to bars in New York, NY in March through July 2019. The COVID-19 pandemic actually strengthened this effect – in March through May 2020, a 10° F increase in the AccuWeather RealFeel Temperature led to an 8% uptick in visits to restaurants and bars in New York, NY. With restaurants and bars restricted to outdoor seating and drinks to-go, perhaps New Yorkers were more likely to take advantage of these options in warmer conditions. Once New York started to recover and people started moving about again in June through July of this year, temperatures’ impact on traffic went back to roughly what we saw in 2019 – a 10° F increase in the AccuWeather RealFeel Temperature led to a 5% uptick in both restaurant and bar visits.

Sunshine also has a positive impact on traffic to restaurants and bars in New York. Sunny days brought a 6% uptick in traffic to restaurants and a 7% uptick in traffic to bars, compared to days in March through July 2019 when it was not sunny. This response was also exaggerated by COVID-19 – in March through May 2020, sunny days brought a 13% uptick in traffic to restaurants and a 15% uptick in traffic to bars in New York. However, moving into recovery in June and July, sunny days only brought a 9% lift to restaurants and a 10% lift to bars in New York, as people’s behavior went back to normal.

Another key behavior influenced by the temperature in New York is outdoor activities. New Yorkers tend to venture out more as the weather improves in the spring, with a 10° F increase in the AccuWeather RealFeel Temperature bringing a 6% increase in visits to trails in the New York area in March through May 2019. However, moving into the summertime, warmer weather actually seems to make people visit trails less – a 10° F increase in the AccuWeather RealFeel Temperature brought a 3% decline in visits to trails in June and July 2019. With the COVID-19 pandemic in 2020, we saw this trend exacerbated in both directions – a 10° F increase in the AccuWeather RealFeel Temperature led to a 10% uptick in visits to trails in New York in March through May 2020, but brought a 14% decline in visits to trails in June through July 2020.

Beyond temperature, sunshine also has a significant impact on visits to trails in New York. Sun is particularly impactful in the spring – sunny days saw a 20% increase in visits to trails in March through May 2019 and a 16% increase in June and July 2019, compared to days when it was not sunny. The COVID-19 pandemic has strengthened this trend as well – sunny days saw a 46% uptick in visits to trails in March through May 2020, and a 42% uptick in visits to trails in New York in June through July 2020.

Digging Into The Data In Denver
In Denver, Colorado, we see a strong positive correlation between the temperature and foot traffic to outdoor spots like trails, particularly in the springtime. In 2019, a 10° F increase in the AccuWeather RealFeel Temperature led to an 8% uptick in traffic to trails in March through May, and a 3% uptick in June through July. Isolating data from 2020, we see that COVID-19 has exaggerated this trend – a 10° F increase in the AccuWeather RealFeel Temperature drove a 20% increase in visits to trails in Denver in March through May 2020, and an 11% uptick in traffic to trails in June and July 2020.

Similar to New York, we typically see a slightly positive correlation between temperature and foot traffic to bars in Denver. A 10° F increase in AccuWeather RealFeel Temperature led to a 3% uptick in bar traffic in March through May 2019. This trend strengthened during the height of COVID-19 – possibly because people were able to enjoy outdoor drinking experiences as the weather warmed. A 10° F increase in AccuWeather RealFeel Temperature led to a 6% uptick in bar traffic in March through May 2020. This effect was further strengthened by the summertime – a 10° F increase in AccuWeather RealFeel Temperature led to a 6% uptick in bar traffic in June and July 2019, and a 17% uptick in bar traffic in June and July 2020.

Another key trend in Denver? While temperature did not meaningfully affect traffic to restaurants in Denver in 2019, we found a correlation between the AccuWeather RealFeel Temperature and restaurant visits in 2020 – perhaps because people are more likely to be eating outdoors as the weather improves, due to the pandemic. A 10° F increase in the AccuWeather RealFeel Temperature led to a 7% uptick in restaurant traffic in March through May 2020 and an 18% uptick in restaurant traffic in June through July 2020 in Denver.

Analyzing Atlanta
In Atlanta, Georgia, non-essential businesses like restaurants and bars began reopening on April 27th, much earlier than in the rest of the country. During these early days of the pandemic in March through May 2020, we found a meaningful correlation between AccuWeather RealFeel Temperature and foot traffic to restaurants and bars in Atlanta, with a 10° F increase in the AccuWeather RealFeel Temperature driving an 8% uptick in visits to restaurants and a 12% uptick in visits to bars. This indicates that people in Atlanta were more likely to venture out for food on warmer days, either picking up takeout before dine-in was permitted, or enjoying outdoor dining after the reopening. However, in June and July 2020, AccuWeather RealFeel Temperature had no significant impact on restaurant and bar traffic in Atlanta, Georgia, indicating behaviors went back to normal. Similarly, there was no meaningful correlation between the AccuWeather RealFeel Temperature and restaurant and bar traffic in Atlanta in March through July 2019, demonstrating that people in Atlanta are just as likely to eat and drink out whether it’s warmer or cooler on days in the spring and summer.

Similar to New York, location data also showed a positive correlation between temperature and foot traffic to trails in Atlanta during the height of the pandemic, as people enjoyed running, biking, and other outdoor fitness activities. A 10° F increase in AccuWeather RealFeel Temperature resulted in a 6% uptick in visits to trails in Atlanta in March through May 2020. However, temperature did not have a meaningful impact on visits to trails in Atlanta in June through July 2020, nor in March through July 2019.

While temperature may play less of a role, sunshine does have a meaningful impact on visits to trails in Atlanta. In March through July 2019, sunny days brought an 18% uptick in visits to trails in Atlanta. This response was exacerbated by the pandemic – sunny days brought a 23% uptick in visits to trails in March through May 2020, and a 21% uptick in June and July 2020. Hence, people in Atlanta tend to enjoy outdoor activities more when the sun comes out in the spring and summer.

Making the Most Out of First Party Data

Quantifying how different forms of weather affect behavior enables data-driven marketers to capitalize on preferences in real-time. Users of the popular AccuWeather app who voluntarily opt-in to receive local weather forecasts also have the advantage of maintaining control over when AccuWeather and its third-party partners may access their data.

Brands should take notice, combining weather data with location-based insights to:

  • Analyze how different types of weather impact foot traffic to your locations. While people may behave a certain way when it comes to the category overall, people may respond differently for your particular chain (for example, last year we showed that steakhouses actually saw fewer visits on unusually hot days).
  • Target specific location-based audiences at the right place in the right moment with AccuWeather. For example, marketers might identify and reach frequent bargoers using the AccuWeather app or AccuWeather For Business insights on warmer days, when they’re more likely to visit a nightlife spot.
  • Measure the ROI of those campaigns in terms of incremental foot traffic, comparing performance across different targeting strategies, creative, and more.

What’s Next?

Marketers no longer need to make assumptions about how weather may affect foot traffic, as there are accurate datasets available that can provide real world data and insights, such as weather-based analytics and Foursquare’s proprietary foot traffic data. With the prospect of COVID-19 cases trending upward during the colder months, smart retailers and QSRs can use these insights to estimate changes in consumer behavior and make the appropriate adjustments to their operations, including supply chain, staffing and delivery. Leveraging the power of these insights, savvy marketers can better understand how both temperature and precipitation affect peoples’ movements—and ultimately, business outcomes.

Interested in custom location-based insights? Email Foursquare at hi@foursquare.com today. Looking for weather-based executions or insights? Email Accuweather at adsales@accuweather.com.

Editor’s Note: Foursquare analyzes foot traffic patterns of millions of Americans that make up our always-on panel. All data is either anonymized, pseudonymized, or aggregated, and is normalized against U.S. Census data to remove age, gender and geographical bias.

For this report, AccuWeather provided historical data for New York City, Atlanta, and Denver from 2019 and 2020, including daily average temperatures, precipitation levels, and minutes of sun. Foursquare then analyzed the impact of weather on foot traffic to different categories, developing a linear regression model to estimate the relationships between foot traffic and weather while controlling for cyclical weekly foot traffic patterns and monthly seasonality. Foursquare additionally measured the significance of the difference between weather/foot traffic correlation coefficients for pre- and post-COVID-19 periods to further understand changes in weather/foot traffic relationships between these periods. Each percentage indicates the average change in daily foot traffic given a one unit change in the predictor variable (temperature in °F, inches of precipitation, or minutes of sun). For this analysis, Foursquare defined sunny days as days on which there were over 100 minutes of sun.