Small Business Survey Scotland: 2023-2024

Methodology

The survey for Scotland is part of a UK-wide large-scale telephone survey of 7,799 SME employers, commissioned by the Department for Business, and Trade (DBT) and was conducted between October 2023 and April 2024. The survey is the latest of a series of annual and biennial Small Business Surveys dating back to 2003. From 2015 onwards, the survey introduced a longitudinal tracking element and is now carried out on an annual basis. The longitudinal element of the survey established a ‘panel’ of businesses that might be resurveyed in subsequent years. This is to allow for analysis of how combinations of factors affect business performance over time. This report provides a snapshot of the state of SMEs with employees (defined as businesses with between one and 249 employees) surveyed in Scotland between October 2023 and April 2024. Enterprises with no employees have been excluded from the dataset on which this report is based. This is consistent with the reporting of the survey from 2015 onwards.

In 2023-24, 764 SMEs with employees took part in the survey in Scotland. 483 businesses in Scotland had taken part in prior year surveys (“panellists”) and 281 businesses were new to the survey in 2023-24 (“top ups”). The survey sample is stratified by business size (micro businesses with 1-9 employees, small businesses with 10-49 employees and medium-sized businesses with 50-249 employees) and by sector (defined by Standard Industrial Classification (SIC) 2007).

The data has been weighted to ensure that the results are representative of the overall Scottish SME population. Because of changes in terms of sampling and changes to the questionnaire in 2023-24, data cannot always be compared with previous editions of the Small Business Survey Scotland.

This report contains selected findings. The accompanying data tables show the full findings for Scotland across the following themes:

• business demographics / characteristics
• trade activities
• business practices
• access to finance
• performance and outlook
• energy efficiency

These are broken down – where sample sizes allow – as follows:

• business size (micro, small, medium)
• sector
• location (urban or rural)
• women-led (women-led, equally led, women in minority, entirely male-led)

The women-led breakdown has been added for the first time to the findings report for the Small Business Survey Scotland, and accompanying data tables, in response to demand from stakeholders for more evidence on women-led businesses. It is anticipated that this will continue to be provided in future releases.

This survey differs from the Business Insights and Conditions Survey (BICS) in that the Small Business Survey includes firms with fewer than 10 employees, thus is more representative of the Scottish SME base as a whole. While BICS is published fortnightly and provides the most up to date results, this annual survey generally asks for more details and a longer-term view with its questions.

Table 1: Sample Breakdown by Sector, Longitudinal Small Business Survey 2023, Scotland

To reduce the average interview length and therefore boost response rates, three cohorts (A, B and C) were utilised for the LSBS 2023 survey (Table 2). This year, all the respondents were randomly assigned to one of the three cohorts (as opposed to last year, where some survey respondents did not feature in a specified cohort). Some questions in the survey were only asked of one cohort, but the majority were asked of all survey respondents. Throughout the report, it is made clear where questions have been asked to one of the cohorts only or the full sample.

Table 2: Sample Breakdown by Cohort, Longitudinal Small Business Survey 2023, Scotland

Comparisons drawn between sub-groups (e.g. Scotland vs. UK; 2023-24 survey results vs. 2022-23 survey results) in the report are displayed only when the difference with the overall finding is statistically significant at 95 per cent confidence level. Due to smaller sample sizes for sub-groups, the survey's estimates may be affected by sampling errors and therefore apparent differences of a few percentage points between sub-samples may not reflect real differences in the population.