Wrapping up ‘Black Friday’: How the ONS captures the effect of a major shopping trend

It’s an American import that in just a few years has changed our traditional shopping habits in the pre-Christmas period. Now it looks like ‘Black Friday’ is here to stay.  Rhian Murphy explains how the ONS takes account of the impact of  ‘Black Friday’ when compiling the UK’s most comprehensive and influential retail statistics.

Earlier in the year, I wrote a blog on the fast changing retail sector, which looked at the growth in online sales.

Another notable change in the retail sector has been the evolution of Black Friday, which first made its mark in the UK back in 2013 as a one-day promotional event for Christmas. Since then the phenomenon has evolved with retailers introducing longer sales periods, sometimes spanning over two weeks.

To add to the changes with this promotional event; this year in 2019, the official ‘Black Friday’ lands on the 29th November and falls outside of our usual November reporting period for the official retail sales estimates.

We currently collect data from retail businesses on a regular pattern, sometimes covering 4 weeks and at other times 5 weeks, as this is how many retail businesses report, or used to report, their results. However, this will change in the longer term as we review our approaches and look to introduce new sources.

In 2019 our estimates for November cover a four-week period from 27 October to 23 November, meaning that only promotions from within these dates will be included in November, with Black Friday falling within the December figures. Data from November’s retail sales will also feed through to November GDP.

So, what does the timing of Black Friday mean for ONS and the Retail Sales output?

We have an approach known as ‘seasonal adjustment’ that allows us to adjust retail sales for the timing of events such as Black Friday to give us the best picture of what is happening in the shops. As the impact of Black Friday has changed over time it can also be helpful to look at the November and December data together.

To ensure we have the best estimates for retail activity, we have completed an additional exercise this year to engage with retailers and gather further commentary on their Black Friday promotions. We know that various stores can behave differently and any information on their activity will help inform our analysis. We know, for example, that some stores do not engage in Black Friday promotions, while others will take full advantage.

We closely analyse all the data and have a dedicated methodology team that can apply statistical methods to account for any changes in activity, if necessary. Our analytical approach provides us with the flexibility to treat each sub-sector individually to account for any difference in consumer and retail behaviour when it comes to seasonal promotions, such as Black Friday and Christmas. We can apply unique treatments for specific store types, which is a great benefit to us.

Our wealth of data from approximately 5,000 contributors of all store types and sizes, in addition to our long span of consistent data – running back to 1986 – means that we are well placed to understand and manage these changes. This is important since our retail estimates form a crucial part of our wider measures; the Index of Services and GDP.


Rhian Murphy – Head of Retail Sales at ONS