Research and analysis

VOA non-domestic property attribute data: Quality Assurance of Business Floorspace Statistics

Published 17 December 2024

1. Executive Summary

The Valuation Office Agency (VOA) has been undertaking a program of work to review our business floorspace statistics. Whilst the VOA has been publishing statistics on business floorspace since 2016, it is problematic to produce due to the complexity of the data. To address this, we needed to understand the data in more detail and have taken the decision to review the methodology, documentation, and quality assurance to ensure the statistics best meet user needs. As part of this review, we have deemed it useful to publish this report as a more comprehensive description of the quality assurance of our non-domestic property attribute data. The report is purely detailing the quality of data with regards to use within official statistics and does not reflect the quality of the data for valuation purposes.

Floorspace data is a vital component in the majority of non-domestic valuations carried out by the VOA. It is also an important statistic as it provides information about how non-domestic properties are being used and could therefore be seen as an indicator of economic activity.

This report has been published to inform a user consultation on the business floorspace statistics, and we invite all users to provide their feedback. We hope that the information presented below helps to explain the decisions we have made and can inform users in their response.

2. Introduction

The assessment of this data source is carried out in accordance with the Office for Statistics Regulation’s Quality Assurance of Administrative Data (QAAD) toolkit. This strives to give a proportionate approach to assessing the level of quality assurance required for the data source and methods used for producing statistics on business floorspace in England and Wales.

The VOA is an executive agency of Her Majesty’s Revenue and Customs (HMRC). It has been responsible for maintaining the rating lists for Non-Domestic Rates in England and Wales since Non-Domestic Rates was first introduced in 1990.

This report presents key research into the quality of the non-domestic property attribute data collected by the VOA. Through this document we aim to provide information and assurance to users that property attribute data collected by the VOA are suitable for the purpose of producing reliable statistical outputs. It will describe the strengths and limitations of the data and raise awareness of how it is currently utilised in the agency.

The report is structured as follows:

• Section 3 describes the Quality Assurance of Administrative Data (QAAD) Toolkit and summarises our assessment of assurance level required for the data source
• Section 4 outlines the operational context of the admin data collection
• Section 5 covers communication with data supply partners
• Section 6 shows the quality assurance principles, standards and checks applied by the data supplier
• Section 7 looks at our quality assurance investigations and documentations
• Section 8 discusses and summarises the overall strengths and limitations

3. Assessment of quality assurance level

The (QAAD) Toolkit sets out 4 levels for the quality assurance that will be required of a dataset. The assessment of the assurance level is in turn based on a combination of assessments of data quality risk and public interest. The toolkit sets out the level of assurances required as follows:

Assurance level	Description
A0 - No Assurance	Not compliant with the Code of Practice for Official Statistics
A1 - Basic Assurance	Statistical producer has reviewed and published a summary of the administrative data quality assurance (QA) arrangements
A2 - Enhanced Assurance	Statistical producer has evaluated the administrative data QA arrangements and published a fuller description of the assurance
A3 - Comprehensive Assurance	Statistical producer has investigated the administrative data QA arrangements, identified the results of independent audit, and published detailed documentation about the assurance and audit

For the purposes of using VOA data to produce statistics on business floorspace, the official statistics team considers that there is a medium risk of quality concern and a lower public interest, so an enhanced assurance (A2) is applied.

The level of risk of data quality concerns for producing official statistics is medium because:

• Although data have multiple sources including information submitted from different local authorities, the planning portal, building developers, ratepayers and inspections; the data supplier team is located within the same organisation as the statistics producer team, meaning there are effective communication arrangements.
• The data are manually input by valuers across multiple teams both at a national and regional level, however there are separate quality assurance and verification processes.
• Data are collected for the purpose of valuation rather than statistical outputs.
• There are known limitations with the data but processes are in place to exclude data that is not appropriate for statistical production.

The official statistics production team considers that the level of public interest in floorspace statistics is low because:

• The data will be used to produce specific statistics on business floorspace in England and Wales. In the past these statistics have contributed to policy or decision making in a very limited way.
• The statistical outputs get minimal media coverage and are of a low economic and political sensitivity.

It is important to note however, that the statutory use of floorspace data is vital in valuing non-domestic properties. We also perceive an interest in record level floorspace data and we share this information with other government departments.

The following sections outline the considerations taken in order to provide assurance that the VOA’s non-domestic property attribute data are of sufficient quality to produce official statistics on business floorspace.

4. Practice area 1: Operational context and admin data collection

Data Collection

Floorspace data are now largely acquired by the VOA through multiple sources including local billing authorities, the planning portal, building developers, ratepayers and online resources. Previously, inspections were the main method of data collection, however, in recent years the VOA has reduced the number of inspections deemed necessary and replaced these with digital research. It is worth noting however, that inspections have increased again for the 2026 revaluation.

There are many sources of information for non-domestic properties, as their valuation methods are diverse. Some of the methods used to collect and update data are:

1. Information from ratepayers or their representatives who can update property information through the Check system.
2. Third party sources: information that is available to the public, for example local authority planning records on the planning portal or individual property records held by private companies.
3. External inspection of the property where the relevant details visible from the road can be confirmed. For example, seeing the property type or the presence or absence of alterations.
4. Internal inspections of the property where attributes can only be confirmed internally, or a re-measurement is required.

Legal basis for collection

It is a legal requirement for the VOA to maintain accurate rating lists for business rates under the Local Government Finance act 1988, which necessitates collection of floorspace (and other) data to enable trusted valuations. The implementation of this function has been informed by subsequent legislation and case law. The VOA makes every effort to ensure the data are accurate.

Original purpose for data collection

VOA capture data about properties to compile and maintain rating lists, specifying a rateable value (RV) for each relevant non-domestic rateable property (also known as a hereditament) in England and Wales. These rateable values provide the basis for national non-domestic rates bills, which are issued by local billing authorities.

Triggers to collect new data

When a new property is built, the attribute data are collected and entered into the database as a new entry. Attribute data for existing properties are only updated when the VOA is made aware of inaccuracies or amendments. There is a reliance on local authorities and ratepayers to inform VOA of changes to properties, at which point VOA will collect the updated attribute information digitally or via inspection if this is deemed necessary.

If a ratepayer makes alterations to their property or believes the information is incorrect, they can submit a Check through the VOA’s Check Challenge and Appeal process. They will do this directly or through an agent. The Check process allows ratepayers or agents to confirm the details of their property, and the VOA will update the information they hold. However, the Non-Domestic Rating Act 2023 states that ratepayers will be legally required to notify the VOA of changes to the occupier or physical property characteristics, and to provide rent and lease information to the VOA. This new duty to notify will be phased in between 2023 and 2026 and will replace the Check process described above.

When Billing Authorities are made aware of a change that could impact RV, they are legally required to notify the VOA. A process is in place where billing authorities can submit a report that will allow the VOA to update the information about a property.

VOA staff also have a statutory duty to compile and maintain accurate local rating lists. As part of this duty, valuation officers are required to note any physical changes uncovered during inspections or research and submit a report to make an alteration to the list, updating the information about the property in the process.

Potential sources of bias and error in the administrative system

As the purpose of data collection is for valuing properties, the information that is collected by valuers is essentially that which is needed for valuation. There is a large amount of variation in non-domestic properties and a variety of valuation methods. Not all valuation methods use floor area as a basis for valuation and there are some instances where the floor area variable does not store an area measurement but rather a number of units (for example the number of rooms in a hotel rather than the floor area of the hotel).

Revaluations

The agency regularly undertakes a revaluation of all properties on the rating list. It is currently legislated that this will happen every 3 years. Revaluation will result in relatively little maintenance of property attribute data, although the VOA may request updated rent and lease details for properties. However, the increase in frequency of revaluations is raising the profile of property attribute data in the agency and more inspections are being carried out for the 2026 revaluation to improve data accuracy.

5. Practice area 2: Communication with data supply partners

The data received by the producers of the statistics have gone through a well-established data pipeline. The raw data are recorded by valuers in line with agreed upon standards for valuation. Data engineers and IT specialists then present this data in easily accessible curated data sets that can be used by statistics producers and analysts across the agency. Collaborative relationships between all relevant teams exist, ensuring data quality issues are identified and acted upon.

The data are initially collected by Valuers and recorded in the Agency’s Central Database. Since the producers of official statistics are of the same organisation (VOA) as the data collectors (Valuers) there is frequent and open communication. This allows for an easy exchange of knowledge around the operational processes that might affect the data quality and production of official statistics.

The data used by the producers of official statistics have been curated by IT specialists and Data Engineers in agreed upon and established data pipelines. Known data quality issues are acted on and communicated to all users of the data. If data are missing or the pipeline has failed, this is acted on quickly and communicated to all users. There is a minimal risk that the data are not available to producers, and there are open channels of communication were anything to go wrong in the supply of data.

6. Practice Area 3: Quality Assurance principles, standards and checks applied to data supplies

Approach to data collection and quality

As described in section 4, the VOA collect property attribute information for the purposes of maintaining the local rating lists. The VOA maintains the rating lists, updating information about properties, adding new properties and removing properties when needed.

In principle the VOA will collect and update information in line with the Rating Manual, a technical guide for Valuation Officers, that represents their understanding of the law. It provides guidance on valuation methods for different types of non-domestic property. This will inform the type of information collected and how property attributes should be measured for different types of business property.

For a large majority of properties, the floorspace and usage of a property are required to be recorded to submit a valuation, though there can be variation in how this is recorded. For some specialist properties other attributes or information about business activities are recorded for valuation.

Floor Area Measurement standards

In line with the rating manual and based in the Royal Institution of Chartered Surveyors (RICS) Code of Measuring practice, there are two main methods used to measure floorspace: Net internal area (NIA), which measures only the usable area measured up to the face of perimeter walls, and Gross internal area (GIA), which measures the whole enclosed area of a building, including internal walls, columns, chimney breasts etc. Therefore, properties measured using GIA will have a slightly higher floorspace measurement than if the same property was measured using NIA.

RICS’ practice states that NIA should be applied to the measurement of Offices and Shops (excluding hypermarkets, Superstores and Retail warehouses) whereas GIA should be applied to Industrial, Storage, Distribution, Hypermarkets, Superstores and Retail warehouses.

QA Checks by Data Collectors

Following collection of data by valuers, the data attributes are input manually into the VOA central operational database (CDB) and other digitalised systems. This input is subject to quality checking by the individuals inputting the data and, in the case of CDB data, undergoes a number of validations which are built into the software recording system to ensure the properties’ attributes are consistent with other comparable properties and have the required information for the valuation.

As non-domestic property valuation is diverse and complex, there is a reliance on senior and more experienced valuers to spot potential errors in the data. These valuers will be subject matter experts in particular types of properties and work both as managers of regional valuer teams and in separate audit teams.

QA Checks in data pipeline

IT specialist and Data Engineers consistently monitor and develop the data pipelines. Automated processes are in place that improves the quality of the data and checks for errors. This includes adjustments to un-validated data fields and daily error checking. Dedicated teams are interrogating any changes in the data and continuously improving processes.

Implications for Official Statistics

Standards of data collection are tied to the legislation governing valuers as laid out in the Rating Manual. These standards should ensure a consistent approach to data collection and entry, and for most properties should change very rarely. The data has QA checks by both the data collectors and by the specialist data engineers that supply the data. Whilst there are known quality concerns, discussed in section 7, overall, the data supplied to the producers of official statistics is considered sufficiently maintained and monitored, and so appropriate for the production of official statistics.

7. Practice area 4: Producers QA investigation and documentation

Producers

The production of VOA statistics is completed by dedicated teams of Analysts, applying a well-established process. The data are taken from curated data sets intended for analysis and processed in automated pipelines built using the subject matter expertise of VOA analysts. The statistics are compiled, and content is quality assured by a responsible statistician and VOA’s statistics profession lead.

Availability of Data

The availability of property attribute data is dependent on the method of valuation used. Most properties in the Rating List are valued on a rental basis and fall into one of the bulk classes (Shops, Offices, Warehouses, Factories, Land, Miscellaneous). The VOA will gather information on rents paid for a type of property in an area and will work out a price per metre for each floor area. We estimate that about 90% of all non-domestic properties are valued on a rental basis. If a property has been valued on a Rental basis it should be expected that a measure of floor space has been collected and recorded on the Central Database (CDB). There are some exceptions to this, as detailed in the “potential sources of error and bias” section above.

The other 10% of properties have been measured in one of two specialist ways: the Receipts and Expenditure (R&E) Method, where they are valued in simple terms of how much money they make, for example pubs; and the Contractor Basis ^{[footnote 1]}, which is the cost to rebuild the hereditament, for example hospitals. These properties are classed as non-bulk, and their attribute data are neither uniformly recorded nor easily accessible and so floorspace data for these properties are not suitable for producing official statistics. Given this, the following discussion will focus only on the 90% of properties that have been valued on a rental basis and have information stored on the CDB.

Structure of Data on the CDB

For the purposes of valuation, the VOA will identify properties as hereditaments. A hereditament is a unit of rateable occupation, broadly defined as a contiguous space with a common occupier. This is a legal definition set out in case law and as such liable to challenge and therefore a degree of change. It may differ from other classifications of properties such as addresses or Unique Property Reference Number. A hereditament is not the same as a building and it is not generally possible to identify buildings from the data. A single office block may contain multiple hereditaments or a holiday complex with multiple cabins may be classed as a single hereditament.

The valuation of a hereditament is based on the survey unit which often constitutes multiple survey lines. A survey line is a subsection of a hereditament, and each is given a value largely based on floor space and usage. These are then summed up to hereditament using suitable multiplicative adjustments. Both hereditaments and survey lines are given a classification; it is at the survey line level that floor space is recorded (manually) for valuation purposes.

Figure 1

Floorspace statistics are currently based on the hereditament. This means that the total floorspace recorded for a hereditament, i.e. the sum of the floorspace recorded against each survey line, is attributed to the overall classification use for the hereditament, regardless of the actual usage of each individual survey line.

Classification of properties and survey lines on the CDB

Non-domestic hereditaments have multiple classifications applied to them. When a hereditament is valued, it is given a Special Category Code (SCat). This is an internal descriptor that primarily denotes how a property is valued. The hereditament is also assigned a Primary Description code. This generally reflects the primary use of the hereditament. There are exceptions to this and if the default Primary Description code is at odds with the use of the hereditament, a free-text, non-standard description can be added.

Special Category Codes

SCat codes are given to all hereditaments recoded on the CDB. It is a system of coding that groups hereditaments according to the method of valuation used.

SCat codes are currently used in our official statistics to group properties into sectors: Office, Industry, Retail and Other. We publish tables on the stock of properties ^{[footnote 2]} (total number and rateable value) for each SCat code.

There are currently 384 SCat codes in operation with 75% of properties assigned to only 10 SCat codes. The remaining 25% of properties are split between 374 scat codes with 152 SCat codes having fewer than 100 properties. 

SCat codes can be subject to change over time. They can go in and out of use and occasionally be reused for a different purpose. This reflects the VOA updating the way they categorise certain properties and organisation of work. Any change would take place at the point of a revaluation. In any case, when floorspace statistics are published, there is a complete time series produced (so while there may be small revisions compared with previous years’ releases as a consequence of SCat code changes, the latest release will be consistent over time).

Analysis found that changes in SCat codes and their descriptions occurred for small numbers of properties, and SCats with the largest number of properties showed relatively small changes year on year. However, this could potentially be problematic when producing a timeseries if there are changes in the way certain properties are categorised.

SCat is a VOA-specific identifier that describes the valuation scheme used to value that property. As such it may not necessarily represent the actual use of the floorspace intended as official statistics. Whilst in the large majority of cases the description of the SCat code would be consistent with that property’s usage, there are instances where this is not the case. For example, it has been found that gyms in office blocks are also valued under the 203 Office SCat code and therefore given the SCat description of office. This is an important limitation for official statistics when using SCats to group properties.

Importantly, the above features of SCats are only of concern for official statistics and do not reflect on the quality of the valuation of properties. The VOA will ensure that all properties are placed in the correct valuation scheme and the correct information collected for an accurate assessment of RV to be calculated.

Primary Description

A separate, additional primary description code is given to all hereditaments on the CDB. This is a three-character code that is intended to reflect the primary usage of a hereditament, although this is not always the case as it is driven by the SCat code and may therefore also reflect the valuation method.

Operational caseworkers advise that more attention is given to ensuring the SCat code is correct as properties are grouped by SCat when delivering a revaluation and this has been the case for both the 2023 and 2026 revaluations. With this in mind we feel that SCat is the more reliable classification to use within official statistics.

Survey Line Use Codes

Survey line use codes are a classification given to survey lines, distinct units of a hereditament that make up part of the valuation. They are intended to represent the usage of that survey line, for example KTN (Kitchen), and will indicate to the valuer the price per m2 that the survey line should contribute to the overall valuation. They are assigned to all survey lines recorded on the CDB.

These use codes are general codes that can be assigned to survey lines in many different types of properties. For instance, ‘OFF – Office’ will not only be assigned to survey lines linked to hereditaments categorised as Offices, but also any other category of hereditament (Shops, Workshops etc.) that contain some office space. These survey lines should be a distinct space with a different and specific use from surrounding contiguous areas.

These codes are stable over time and are rarely changed, though they can be subject to the interpretation of valuers and may not always be recorded consistently.

Quality of floor space data

An area variable is required to be recorded against all survey lines for the valuation to be valid, thus all survey lines on the CDB have an area value recorded. This area variable would be the source of information for publishing official statistics on business floorspace, however, there are options as to how this can be aggregated for statistical purposes as detailed in the public consultation document. The official statistics team has performed analysis and held workshops with valuers to understand all the quality considerations when using this variable.

A key consideration when producing statistics is what can be recorded in the Area variable. When entering information into the CDB valuers can record either an area or unit value. Therefore, when aggregating categories of property any summation of floorspace risks being a combination of unit and area measurements, which would lead to misleading and inaccurate statistics.

In principle, whether a survey line is recorded as a unit variable should be laid out in the rating manual standards for valuation and measurement. It can then be identified in the data and removed from any floorspace counts. For example, ‘Car park spaces’ and ‘Kiosks’ are known survey lines that are recorded on a unit basis and should not be included.

However, there is a wide range of survey lines present in the data going back over time, and how area or units are recorded, whilst dependent on the rating manual, can be affected by the valuer’s discretion, regional variation and human error. It has been found that within a use code that is valued on floorspace, there may still be survey lines that appear to have a unit value recorded. VOA analysts have developed further methods to identify these outliers in the data as best as possible. Review of the distributions of both the area variable and rateable value per metre squared indicates these methods have shown to be effective and acceptable tolerances for outliers will be identified and communicated alongside any future publications of floorspace statistics.

8. Conclusion

Considering the wide variety of property types and information being collected, it is a considerable strength of the VOAs property attribute data that how information is collected is laid out in the rating manual and governed by law, ensuring standardisation around the data collection process.

However, there may still be inconsistencies in the way property attribute data is collected due to the ways different types of properties are valued. For the 10% of properties where the valuation is not based on floorspace it would not be possible or reasonable to publish floorspace statistics. As such VOA floorspace statistics should not attempt to represent the whole of the non-domestic property market, but rather select specific areas that have good data coverage and are in the public interest.

The floorspace data recorded by the VOA has some limitations due to the nature of the central database meaning area and unit information is stored in the same place. However, these limitations can be overcome with the use of VOA specific domain knowledge and the exclusion of outliers. As such for specific survey lines and property types, floorspace can be considered of an acceptable quality to publish official statistics.

There are different options for the way properties are classified within floorspace statistics depending on the intended use of these statistics. If users are interested in floorspace statistics aggregated at a hereditament level, then it is sensible to classify by SCat code and aggregate accordingly. If users are interested in utilising floorspace statistics as more of an economic measure, then floorspace can be classified at the more granular survey line level and aggregated according to the use at this level. More details around these options can be found in the public consultation and we invite users to provide their feedback via the survey.

1. For more information on how different properties are valued go to: https://www.gov.uk/guidance/how-non-domestic-property-including-plant-and-machinery-is-valued ↩

2. Official statistics on NDR stock of properties can be found at: https://www.gov.uk/government/statistics/non-domestic-rating-stock-of-properties-2024 ↩