This 2021 edition of the UK Retail Rankings follows the same format as last year. The report is really about the data and that is included in the Databook. The aim is to include all the major retailers that we can get data for as well as a few that we can’t. In practice that means that we are constrained by the small company provisions at Companies House, which allow companies only to file a brief balance sheet summary, and by the structure of companies. We have no data on partnerships (such as Slater menswear) or on retailers that publish accounts by store (such as Anne’s Cottage)

We follow the change made in 2018 edition of the Retail Rankings when we subdivided the non-store retailers into more meaningful groups and we combined the two electricals sectors in the ONS data.

• The impact of COVID-19 on UK Retailing.

• An overview of mergers and acquisitions as well as administrations in the UK retail sector.

• Major players and developments by UK retail sector.

• A comparison of retailer performance by sales, profit and store data for the leading UK retailers. (Databook)

• Index of trading names and company names for the leading 600 UK retailers. (Databook)

The cut-off point for inclusion is, notionally, sales of £6 million. But in practice the limit is set by the Companies Act 2006, which allows smaller companies to file only very limited accounts. For the smallest companies only a simplified balance sheet is required and small companies do not necessarily need to produce audited accounts.

The thresholds have been steadily increased, most recently in January 2016. In this report we include what data we can find. But we think there must be many companies, especially internet pureplayers, which may have substantial sales but which are able to claim a small company exemption. For example, Chitter Chatter, on online mobile phone retailer had sales of over £40 million in 2016, but was able to file limited accounts in 2017 and 2018.

To qualify for an exemption companies need to fall under two of the three limits shown in the next table. Note that partnerships do not need to file accounts either.

There’s another problem this year, that there has been a temporary extension to the time allowed to file accounts at Companies House. Private companies and LLPs (Limited liability partnerships) have been given an extra three months and so have a full year to file. That has meant that we have had to estimate many more sales figures than usual.

Sales density or sales per square metre is one of the most important metrics we use in retail analysis. Where we can find sales area data, we include it. We also put it in if we think we have a reasonable indication of what it is. But in all too many cases it is just not published and we are not prepared to estimate it unless we have at least some evidence to back the figures up.

The classifications used are in accordance with the latest classifications used by the ONS – the SIC 2007 (Standard Industry Classification).

We deviate from this in a few cases. For example – we have sales figures for both food and non-food for Marks & Spencer and, while technically all should be included in clothing retailers, for this report we split them and treat M&S Food as a food retailer.

We also have a department store classification. This does not exist in the ONS classifications at all. A department store is easier to understand conceptually than to put a definition to. In practice we think that some (eg Debenhams) are classified as clothing retailers and some (eg John Lewis) as mixed goods retailers.

Throughout the report, rankings are computed on UK retail sales for 2019/20 (financial years ending between July 2019 and June 2020 – see below). We try to exclude any non-retail and non-UK sales where they are a significant proportion of the total, but that is not always possible.

There are a few cases where we have had to estimate 2019/20 sales. This can happen in several ways, the most common is that retailers with a June year-end have not yet filed accounts at Companies House (see above in qualifications for inclusion).

There is also a problem with retailers that have gone into administration. Sometimes there is information in the administrator’s report, but often a slimmed down company survives, not always under new management. When that happens there can be a gap in periods reported and we have to put in an estimate.

There are also a few cases where we have had to estimate the whole sales series. This usually happens where the company publishes little or no data (eg Lidl) or where the retail arm is a small part of a much larger group and its results are not split out (this is the case with the retail operations of the oil and mobile phone companies). We include estimates where the company is clearly an important player in the market and to exclude it would be misleading. But we do not estimate where companies are relatively small players in the market and do not file any accounts at all.

Another exception is partnerships, which are not required to file accounts either. Therefore companies such as Maidenhead Aquatics, Burncoose Nursery or Slater Menswear cannot be included. Slater is particularly regrettable. We think that it is a small, but significant player in the menswear market. It trades from cheap sites, such as the first floor of high street stores, and focuses on formal wear. It has 26 outlets, but it is unlike any other significant player and therefore virtually impossible to estimate.

Decisions on whether to put in pro-forma numbers to allow for major acquisitions, or whether to allow the tables to reflect the situation obtaining during 2018/19 are inevitably subjective. In this report we have decided that we should attempt to show the structure of retailing going forwards. So we include a pro-forma figure even when the two companies were trading independently throughout the Rankings year for this report.

ALL COMPANY SALES ARE QUOTED EXCLUSIVE OF VAT. Market/sector size figures are also quoted exclusive of VAT.

The column headed 2019/20 includes all year-ends between 1.7.19 and 30.6.20. As most retailers have years ending in December, January or March, this means that the column headed 2019/20 approximates to the outcome for the 2019 calendar year and allows for proper comparisons between companies with differing year-ends. Unfortunately it also means that the Debenhams financial year to August 2019 and the Dixons Carphone year to April 2020 appear in the same column – 2019/20. No solution is perfect to this problem; we think that the one adopted is the best compromise.

In the notes we give the actual year-end to which the note refers. Notes without a date refer to factors that are true for every year.

Sales are as reported in the accounts. We have made no adjustment for 53 week years, although where a large company has made a major takeover or there is a non-standard accounting period we have estimated an annualised figure. Where this has happened, there is a note to flag it. However, we only include annualised data for the latest year, so that the overall ranking is correct. Previous years show data as published. We do not try to annualise profit figures.

Note that while we think it is essential to estimate sales for the UK Retail Rankings, where necessary, we never estimate profitability.

We quote both operating profit and pre-tax profit:

Operating profit is trading profit after normal operating costs, but before interest and exceptional items.

Pre-tax profit is operating profit less net interest payable and exceptional costs, but before tax.

We have tried to exclude all non-UK and all non-retail turnover were it is significant. However, this is not always possible. For example, it is often not possible to identify sales generated in Ireland, which many companies include in the UK data.

Inclusion of non-retail turnover is a bigger problem – it can lead to some highly distorted numbers for food wholesalers that also run shops, for some of the bakers and for fashion companies that also sell through third parties. Anomalies of this sort are explained in the accompanying notes.

The biggest problems arise in comparing company turnovers with the ONS retail sales data in order to estimate market shares. The ONS excludes several areas of non-retail turnover from its data and it is not possible for us to make any more than a rough estimate of the importance of such categories to the companies themselves.

The major exclusions by the ONS are:

• Petrol sales – over 15% of the sales of the top four leading food retailers

• Trade sales – or business-to-business turnover. This is an important element of the DIY retailers’ sales

• Restaurants

• Income from NHS services

• Sales of services by the mobile phone retailers.

• Car parts.

Where a retailer publishes a sales breakdown we use that, otherwise we use subsidiary accounts where possible, but these should be treated with caution. Balance sheet figures can be deceptive as holding companies can move funds around. Internal accounting, and the technicalities of ownership and putting accounts together, can also distort profit numbers. Subsidiary accounts often give different data to that in the consolidated accounts. For example, the accounts for Waitrose do not match the breakdown in the John Lewis Partnership accounts.

Sales per sq m and sales per outlet are computed on the average size/store number during the year. We believe that this is the best way to remove the distortion from a rapid opening/closure programme. We have, otherwise, made no attempt to adjust for the timing of openings. Nor have we adjusted for internet or mail order sales if the non-store sales are not published.

Sales per sq ft can be calculated by multiplying the sales per sq m figure by 10.76.

Sales area figures are mostly as quoted by the retailers themselves. The vast majority quote them net, ie they exclude storage and office space, and on the rare occasions when we estimate them, they are estimated on that basis as well. However, there can be problems, for example with garden centres or DIY stores, with the treatment of outdoor areas. UK retailers are not consistent and it is a pity that they do not adopt the German practice of including a fixed proportion of external sales areas. In the vast majority of cases we do not try to estimate sales areas if they are not published.

In general we have followed the ONS classifications, though not exclusively. Wherever possible we have calculated market shares for the leading players in each sector. Where we follow the ONS retail classification we use the ONS sales data. But there are several occasions where we believe that a sector is large enough to warrant separate analysis and there is no appropriate ONS sector. In those cases we usually use consumer spending data (as, for example, with Jewellers). In all cases we have indicated which data we have used.

The market size figures used for the charts in the sector sections are sales by specialist retailers. So retailers are compared with their peers. No market share estimate is ever entirely satisfactory. In clothing, for example, the sales of clothing that go through grocers or department stores do not impact on the market share numbers. However, this is a report about retailers and not about spending patterns. Therefore our estimates indicate how the specialists have fared relative to their peers and how the pecking order among them has changed.

Sector names have been abbreviated as follows.