CFIB research methodology: how we conduct our research

For more insights or custom data requests, visit cfib.ca/research or email research@cfib.ca.

Download PDF (Full methodology)

CFIB Research Methodology 

Introduction

Research and analysis are core to CFIB’s mission. We have been basing our advocacy on our members’ views since our creation in 1971. Today, as Canada’s largest non-profit organization devoted solely to the interests of small and medium-sized business owners, and with over 100,000 members across the country, we continue to ground our work with direct input from members.

By combining the disciplines of economics and policy analysis, our research work supports advocacy efforts and is conducted largely–though not exclusively–through member surveys. These surveys provide unique, timely, and accurate data on Canadian small and medium-sized businesses.

Each year, our surveys reach tens of thousands of Canadian business owners and span economic, business, labour, social, and other policy issues. Every questionnaire is meticulously designed to ensure an unbiased and objective approach to both quantitative and qualitative insights. Member responses generate hundreds of thousands of data points and tens of thousands of comments, making CFIB one of Canada’s leading sources of small business data.

Examples of our survey-based research include the monthly Business Barometer®, the Your Voice surveys, and our regular cost of red tape and banking research report series.

This methodology guide provides a detailed look at CFIB’s survey processes and practices, including our main survey types, how results are processed, and when findings are made publicly available.

The guide is organized as follows: chapter 1 outlines the composition and representativeness of our membership base, chapter 2 describes the different types of surveys we conduct, and chapter 3 explains how survey responses are collected, cleaned, and analyzed.

1. Membership representation

Distribution of CFIB membership

CFIB represents over 100,000 member businesses across Canada. Our membership reflects a diverse group of independent businesses of all sizes, with the vast majority having fewer than 50 employees but some more than 100 employees. Our members operate in all sectors of the economy, as well as every province and territory.

We link survey responses with member profile data in our database, allowing us to analyze results by province, sector, business size, and more.

The distribution of CFIB’s membership and survey respondents shows that the data collected is close to the true distribution of economic activity across Canada, despite some discrepancies (Tables 1, 2, and 3). Additionally, some sectors are not directly comparable to those used by Statistics Canada, as we have classified them differently from the North American Industry Classification System (NAICS). This is notably the case for the Agriculture and Natural Resources sectors.

Table 1 – Distribution of CFIB survey respondents, SMEs and GDP, Canada, 2024 (%)

Sources:          

1. For CFIB, Your Voice and Your Business Outlook surveys conducted from January 2024 to December 2024, n = 25,958.
2. Statistics Canada, Tables 33-10-0088-01 (province), 33-10-0087-01 (sector), and 33-10-0088-01 (size of business).
3. Statistics Canada, Tables 36-10-0402-01 (province) and 36-10-0401-01 (sector), and Innovation, Science and Economic Development Canada, Key Small Business Statistics 2024 (size of business).

Note: *Utilities and Unclassified are excluded from the calculation.

To make sure the views of our membership accurately reflect those of the small business community in Canada, we occasionally review representativeness by running weighting checks (see section 3 for details). These checks usually show only small differences at the national level, but provincial and sectoral results may show larger variations.

Given these small differences and to stay true to our mission of representing our membership and all small and medium-sized businesses, our survey results are typically presented without adjustments for weighting.

Distribution of CFIB survey respondents         

Generally, our survey participation is not limited to a small, consistent group of members (see Figure 1). Nearly half of respondents answered only one survey in 2024, while 18% participated in two surveys and 9% in three. Just 6% answered four surveys, and only one in five responded to five or more. This rotating pool of participants helps ensure that a wide range of member perspectives are reflected in CFIB’s survey results.

 Figure 1 - Share of CFIB members by number of surveys answered in 2024 (%)

Source: CFIB, based on 38 surveys, 2024, n = 17,890. 

2. Survey methodology: Approaches and types                   

1. Approaches

We conduct most of our surveys online, currently using the Qualtrics platform. To respect our policy of "one member, one vote," we send unique links to password-controlled surveys which can be accessed from laptops, tablets, or phones. 

This approach offers some key advantages:

• Restricting access to active CFIB members, which must be active Canadian-owned independent businesses.
• Ensuring members can only fill out each survey once, providing data integrity, preventing duplication, and securing reliable responses.
• Allowing linking of survey responses to the information from the membership profile database, making the insights more powerful (for example, we can provide survey results by province, sector, business size, etc.).

Participation is always optional—members choose whether to take part.

We also run in-person surveys conducted by territory sales managers, who meet individually with hundreds of small business owners each day.

More specifically, CFIB currently uses four main types of member surveys. The process and sampling strategy for each are described in the following sub-sections.

2. Types

Our Members’ Opinions in-person survey

Since 1975, CFIB has been using Our Members’ Opinions (OMO), a survey with a mix of tracking and ad hoc questions, during in-person interviews conducted as part of annual membership renewals. This ongoing survey starts with a permanent question that tracks small business owners’ most important concerns and has also featured special ad hoc questions on topics such as municipal issues, labour, or government services. It is typically one page and limited to 5 or 6 questions. This survey is best suited to policy or research questions that need high response numbers, require fine geographic or sectoral detail, or are hot-button issues.

Field surveys

In January 2025, CFIB launched a new survey tool designed specifically for use during in-person interviews (similar to our OMO survey). The goal is to gather timely data from both members and prospects on key small business issues. The survey format is very short, with just a few questions and 4–5 multiple choice response options. To date, we have collected thousands of responses on issues ranging from municipal business licensing and property taxes to the Ontario and federal elections and the ongoing Canada-US trade war.

Your Business Outlook (i.e., Business Barometer®) monthly tracking survey

Started in 1987 as an annual tracking survey, Your Business Outlook became quarterly in 2000, and monthly as of 2009. It was also run weekly for six weeks following the events of 9/11 and twice a month early in the pandemic. It tracks small business confidence, expectations, and operating conditions, through a set of permanent questions. This allows comparison of the data over time. The monthly Business Barometer® time series are not seasonally adjusted, as not all national series present identifiable seasonality. Generally speaking, using the year-over-year data points would allow for fair comparison.

The survey is sent at the beginning of each month to a stratified random sample of about one third of the membership. Since 2009, the survey has generated on average 800 to 900 responses monthly, with a corresponding margin of error of about 3.3% for a level of confidence of 95%. Every third Thursday of the month, we release the monthly results via our Business Barometer® reports (Canada, provinces, and industries) at cfib.ca/barometer.

Your Voice omnibus surveys     

Since 2021, Your Voice surveys have been CFIB’s main online omnibus surveys. Their ad hoc questions cover a wide range of economic, business, and policy issues. While generally national in scope, some questions target specific provinces or industries. Most months, Your Voice is sent to all CFIB members who are not receiving the Your Business Outlook survey that month. In 2024, the Your Voice survey collected an average of 2,614 responses each month it was sent, with a corresponding margin of error of about 1.9% for a level of confidence of 95%. You can access results from all Your Voice surveys at cfib.ca/YourVoice.

Special surveys

For important and urgent topics, or for topics needing in-depth analysis, we send ad hoc online surveys either to the entire membership or to a specific province, municipality, or sector.

Some recent examples of how we use special surveys include the Financing Main Street research series, the red tape research series, the business succession series, the Alberta Pre-Election Survey, the Atlantic Municipal report, and our retail community contributions project.

3.    Data processing & analysis

1. Data cleaning

To ensure accuracy, we apply rigorous data cleaning procedures by removing incomplete or inconsistent responses. We maintain the integrity of response counts by removing non-responses ("no answers") from individual questions. Similarly, we typically include “Don’t know/Unsure” or “Not applicable” as answer options, to capture uncertainty. As a result, sample sizes may vary from question to question. The survey sample size reported represents the number of members who responded to at least one question. For numerical responses, we typically identify questionable outliers through statistical analysis and exclude them to refine our estimates.

Further details on sample size reporting are available in the Survey Results Presentation section.

2. Weighting

CFIB survey results are generally not weighted. We are guided by our mission of representing our members and operate using a "one member one vote" system, so each member has an equal say. Our objective is to measure business sentiment among our membership.

Nonetheless, unweighted CFIB survey results are usually fairly representative of the overall business population in Canada due to the large number of responses collected (generally well over 1,000) and the diversity of size, geographical, and sectoral attributes found among CFIB’s 100,000 member businesses. In other words, our survey results tend to quickly converge towards those of the general business population.

In order to validate our data’s representativeness, we periodically perform weighting control tests on survey results, i.e., we adjust to reflect the Canadian small business population. Applying weights to control for the business size, sector, and province of each member usually leads to small changes of +/- 2 percentage points in the national data (see Figure 2). Breakouts (provincial or sectoral results) can be more sensitive, and differences between unweighted and weighted results can vary more widely.

Figure 2 – Select example from September 2023 Your Voice survey  

Source: CFIB, Your Voice survey, September 2023, n = 2,618. Survey question: Which of the following are a serious concern to your business? (Select all that apply)

3. Survey results presentation

Generally, CFIB reports national values, with provincial and sectoral breakouts available as tables or figures, or in an appendix. Sample distribution is also added in appendices in our regular Your Voice and monthly Business Barometer® reports. An asterisk is typically added to survey data points with small sample sizes, for which the threshold is often set at 40 responses. All members’ responses are confidential and reported only in aggregate form. 

For figures such as debt, loans (e.g., CEBA), or costs, we typically report the average or the median to facilitate interpretation and aggregation. However, we typically clean the dataset by excluding questionable outliers, which helps provide a more stable and meaningful estimate.

For table or figure totals that may not add up to 100% due to rounding, we typically add a note at the bottom of the chart instead of adjusting the figures, as part of our commitment to data transparency.

Projections of results to the full small business population

CFIB's projections and estimates are typically based on numerical responses from member surveys and are extrapolated to represent the Canadian small business population using Statistics Canada’s datasets on Canadian business counts with employees. When creating projections or estimates, we often report averages and exclude outliers (see the Data Cleaning section for more details). Some estimates—such as the sales lost by small businesses during the 2024 Canada Post strike—are further adjusted (e.g., weighted) to better reflect the broader small business community.

4. Margin of error: Probabilistic and non-probabilistic surveys

Most CFIB surveys use an online opt-in sample, therefore relying on non-probability sampling. Some surveys use a slightly different approach, such as a stratified random sample (Business Barometer® surveys), or are conducted in person (OMO, field surveys).

We recognize that, technically, extrapolating beyond our membership involves some assumptions, as the sampling is no longer fully random for the entire business population. However, this approach is justified for several reasons:

• The regional and industrial representativeness of CFIB respondents is generally not too far from the distributions of Canadian GDP or number of private businesses (see Table 1).
• The OECD (2003) reported that many surveys based on membership have been shown to provide reliable information on total output or industrial production.^[1] The OECD states that surveys conducted by business associations usually benefit from higher response rates due to closer connections with the respondents.

In light of these considerations, we generally provide as an accompanying reference to most of our survey results a corresponding 95% margin of error to the broader Canadian small business population, which we consider to be a reasonable and methodologically sound approach.

4.    Methodology of popular reports

Index

CFIB often uses indices to track economic trends or to measure comparative rankings.

Some indices are derived from survey results, such as the long-term and short-term indices from the monthly Business Barometer report, which are based on weighted responses of 12-month and 3-month expectations.

Other indices, such as those found in report card reports (e.g., Internal Trade, Red Tape), are built from publicly available data—qualitative or quantitative. These indices may use weighting, depending on the specific methodological approach employed.

Refer to the report in question for a more detailed methodology (and survey questionnaire if applicable):

Business Barometer

Nowcasting and Forecasting using the CFIB Monthly Business Barometer

Red Tape Cost Estimate

Retailer’s Local Contributions

Small Business Workers’ Compensation Index

Business Succession: Assets at stake estimate

CFIB’s Business Financing Survey

Business Enterprise Pulse

COVID-19 SME Debt Estimate

5. Data request and further information

You can access all our recent research at cfib.ca/research and reach us at research@cfib.ca or on x.com (@CFIB_FCEI_Stats)

[1] OECD. (2003). Business Tendency Surveys - A Handbook.