Recently, Analytics Manager Samantha Potter, discussed the first ten days of the second national lockdown and how this had impacted gas usage of some AMR (Automated Meter Reading) device sites across a range of sectors. This week she looks at a more in-depth comparison of the most recent lockdown with the first national lockdown in late March.
AMR devices provide daily, or even sometimes half-hourly, readings to your gas provider, ensuring that you are billed accurately and removing the manual aspects of having to provide a read. Using analytical tools such as Niccolo insight, are able to convert these readings into real consumption. When this is coupled with the AQ (Annual Quantity) of a site, this allows us to see how a site is performing relative to what we would expect for the type of business and the time of year.
The below graph shows the total usage of around 3.5K sites with AMR devices, compared to what would be expected for this time of year (Seasonal Normal Demand, SND), with three- (green dashed line) and four- (solid purple line) week averages to smooth out the data for weather fluctuations. Also on the graph is the temperature anomaly as a dashed red line, showing whether the average temperature for the week was warmer (a positive value) or cooler (a negative value) than expected for the time of year.
The drastic lows of the first lockdown have not been reached again during the second, although there has been a significant effect on overall usage. This has likely been exacerbated by the warmer than average temperatures seen consistently throughout the first three weeks of November, reducing the need for as much gas used for heating as we would expect for this time of year. However, usage compared to SND was already decreasing prior to the second extended national lockdown, likely as a result of the introduction of the tiering system at different times across the UK.
The above graph also clearly shows the relationship between the outside temperature and the variation in gas usage compared to what we would expect at different times of the year. Comparing the weekly usage percentages with the temperature anomaly line shows an almost symmetrical inverse relationship across September and October before the second lockdown was implemented.
Let’s take a closer look at some of the key hospitality sectors that have been most impacted by the national lockdowns:
Two sectors that have performed above their expected usage throughout the majority of the pandemic have been takeaways (dark blue line) and unlicensed restaurants and cafés (purple line). This is possibly caused by sites showing quick thinking and innovation being able to pivot to takeaway only to still serve their customers, and these sectors have shown real resilience throughout this year.
From the above graph, you can also see the time in late June when it was announced that pubs (yellow line) and licensed restaurants (light blue line) would be allowed to reopen and preparation work began, followed by an even sharper increase the weekend they opened at the start of July. This is then followed by a usage peak in the graph throughout August when the Eat Out To Help Out scheme was in full swing, boosting usage in the licensed restaurant sector to 50% above what would be expected at this time of year. Since this point, a steady decline has occurred across all five of the sectors relative to the SND values, with usage plateauing at a minimum over the last few weeks during the second national lockdown.