At a high level, the story hasn’t changed too much over the past week; fuel sales remain down 30-60% from their pre-COVID averages across nearly every market nationwide.
Additionally, the stabilization trend that we noted in our previous post appears to be holding true; the steep declines that we saw in the first few days and weeks following the institution of COVID-related policy measures have tapered off as markets have settled into a “new normal.”
This trend appears to be generally true across most markets nationwide; however, some interesting patterns may be emerging when we segment markets based on size. We segmented the 150+ markets for which we have sufficient data into small (<500K households), medium (500K-1M households), and large (1M+ households) cohorts, and took a look at the change in average daily fuel transactions per site across each of these groups over the past 3 weeks (March 28-April 11).
As the chart below illustrates, the trend is roughly the same across all three cohorts - the relatively steep declines towards the end of March appear to have tapered off in each of the groups.
There are some important differences, however. First, the percentage declines are notably greater in the “large” markets relative to the “medium” and “small” markets. Second, while demand in the “medium” and “large” markets continued to decline over the past 2 weeks (albeit at a slower rate than the previous 2 weeks), the “small” markets actually stopped declining (transaction volume held roughly steady, going from 31.5% declines from pre-COVID totals for the week ending April 4 to 31.4% for the week ending April 11).
It’s too early to tell if this trend will continue, but this could be a sign of demand beginning to bounce back, particularly in smaller / less urban markets.
What happens to GetUpside’s incremental methodology in tumultuous periods like this?
Consumer demand for fuel shifts frequently, both for predictable reasons such as travel patterns around the holidays, and for unpredictable ones such as the current global disruptions resulting from COVID-19. One question we often hear is, “does GetUpside’s incremental measurement methodology still work during periods of time when there is a severe drop-off in demand?”
The answer to this question is yes, GetUpside’s methodology continues to work and accurately measure incrementality even at an unprecedented time like this. Why? Because incrementality is measured against a control group of non-GetUpside users, and the virus affects GetUpside users and the control group in the same way. In the “Measuring Incremental Gallons” document included in this packet, you can read more about how our test and control methodology works and why it’s considered a best-in-class approach.
During this uncertain time, we are closely monitoring the impact of new policies on your business, and making that information, as well as other data that we have compiled, available for your use. By analyzing data from tens of thousands of gas stations across the U.S., we found that steep volume declines predictably follow stay-at-home orders; but we have also found that declines begin to level off after about a week at 50-60% of pre-COVID levels. The remaining volume is largely related to essential travel—whether that’s essential workers traveling to and from their jobs, or consumers making essential trips to pick up home necessities.
The charts below illustrate what happened to demand in two markets—Chicago and Washington, D.C.—during two different periods of demand drop-off. The first period is over the holidays (December 2019-January 2020), and the second is after shelter-in-place orders were enacted to stop the spread of COVID-19 (March 2020). Across both GetUpside users (the top line) and the control group of non-GetUpside users (the bottom line), the average number of visits dropped, but GetUpside was still able to drive incremental visits (the difference between the two lines) in those markets. As noted above, an incremental transaction is one that would not have happened without GetUpside. These visits are above and beyond their control group of similar customers.
While the number of visits from both groups declined, the charts above show that GetUpside is still able to boost users’ purchase behavior above the control group’s behavior, which is what we call incremental. If a transaction is not incremental—as in the GetUpside users (top line) are not transacting above the control group (bottom line)—you do not pay and GetUpside covers all costs associated with that transaction.
Visit our COVID-19 Fuel Impact page at https://app.getupside.com/covid-19-fuel-volume-tracker for regular updates on what we are seeing in your market. You can also find additional GetUpside educational materials in our Knowledge Center at getupside.zendesk.com.