Get in touch
+44 203 865 8689hello@edgepetrol.com30 Newman Street
Chat to us nowClose

EdgeIndex issue 2: Supermarket Showdown, Part 2

Written By

Sebastian Delgado von Euw


May 14, 2019






EdgeIndex (April 2019) 🎉

Heading into the summer we continue to see fuel prices increase, with the national average for unleaded increasing 3.38ppl vs the past month while diesel went up just 2.09. Meanwhile, EdgePetrol stations seem to be faring well maintaining an average margin above 6.50ppl.

EdgeIndex - Supermarket Showdown Volume 2.

Following on our findings from the previous EdgeIndex issue (read it here), this month we seek to evaluate how the same stations behave in terms of volume sold. The question we are asking this week is:

How does the presence of a supermarket petrol forecourt affect nearby non-supermarket petrol forecourts in terms of volume?

Why do we care about volume? 🤔

We have already established that stations close to supermarkets suffer from low margins. But is that always a bad thing?

Not necessarily, if your station has low margins and large volumes you could still be better off compared to a station with better margins and low volumes. Therefore it could be the case that stations close to supermarkets also operate in a larger market with more demand and therefore do not need to match non-SM sites on margin in order to cover their costs.

In order to test this theory we reuse our existing sample of 74 forecourts over the months of November 2018 to February 2019. Within this sample we have:

  • 43 forecourts within 10km of a supermarket. We will call these SM forecourts.
  • 31 forecourts without a supermarket competitor nearby. We will call these non-SM forecourts.

How do these two groups differ in terms of volume?

We again produced a histogram (in appendix) that compared the volume distributions of both groups. Our initial assessment finds that both groups are very closely related in terms of total volume.

The worst performer in terms of volume for both groups is close to 400,000L while the best performing station in both groups is around 3,500,000L. The average total volume sold during our observation period for both groups was ~1.4 million litres. A statistical test also concluded that there was no significant differences between both groups [1].

We looked at three key metrics from our stations: Margin, Total Volume and Gross Profit from fuel. The work showed us that:

  • Gross Profit has a positive linear relationship with Total Volume. In other words, the more litres you sell - the higher your gross profit from fuel will be. This makes the case for investment in higher volume sites, regardless of the competitive nature of the market around them.
  • Higher margins don’t necessarily lead to higher gross profit from fuel. It is also worth considering that, retailers with smaller volumes need to increase margins to make sure they cover the additional costs they lose in economies of scale.
  • Supermarkets do not seem to have a clear cut effect on volume or gross profit from fuel. Volume is maintained at stations close to supermarkets by staying close to them on price, as we discussed in last month’s EdgeIndex.
  • The distance to the closest supermarket yields no clear relationship to volume, gross profit or margin.

Looking further into this we plotted our three metrics against the number of supermarkets close to each station and found a slight negative linear relationship between each of the metrics and the number of supermarkets present. With an increasing number of supermarkets all metrics tend to decrease. This might be a good indication of how competitive their local markets are. But we should take this result with a pinch of salt since we have very few stations that are close to 3 or more supermarkets.

Conclusion 📙

  • Forecourts close to supermarkets show on average lower margins than other stations.
  • But there is no strong evidence that being close to a supermarket means less volume sales or less gross profit from fuel.
  • There is weak evidence that total gross profit from fuel and volume decline as the number of supermarkets close a station increases.

Sebastian Delgado von Euw is EdgePetrol’s Data Scientist. He previously worked in the finance industry and studied Economics at the University of St Gallen with a specific focus on econometrics, financial mathematics and statistics.


  1. T-test results: t-value = -0.18, p-value = 0.86


In this graph each plotted point is one of the stations on our sample. Black dots are SM-forecourts and green dots are non-SM forecourts. The dots allows us to see how each station behaves in terms of different metrics. While the three plot on the diagonal show how each of the metrics is distributed:

  • The number of supermarkets close to a forecourt seems to have a weak negative relationship on all metrics.

  • The only metric where both groups show a significant difference is Margin (which we explore in our previous edition):

  • Higher volumes lead to higher profit:

  • Higher margins don’t necessarily lead to higher  gross profit from fuel:

  • Distance to the closest supermarket does not have a cleat cut effect:

Insights and stories from the EdgePetrol team.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.