About Our Data
Sageworks aggregates private-company financial statements from accounting firms, banks, and credit unions through a cooperative data model with our clients so we can provide up to date benchmarking data in our reports and services. The data is entered by finance professionals, people who understand the basic inputs and mechanics of financial statements, so they are more likely to enter information correctly.
Sageworks takes several precautions to ensure that data entering the database reflects the true operational performance of the companies being analyzed. The data is filtered using six methods:
- Data coming from all Sageworks employees, trial accounts, and clients who opt out of our data collection process is excluded.
- Data marked as “sample” by users or data that must be annualized for our program is removed.
- Financial statements that do not comply with basic accounting principles are removed. (e.g., Net Income cannot be greater than Sales.)
- Data points (or outliers) that fall outside our statistically set bounds are omitted from the calculated averages.
- Duplicate data for a company is removed, with the most recent data taking precedence.
- Statements that have missing fields (such as industry code, location, or time period) are excluded.
Please view our citation guidelines before referencing the data. For more information, read about what we do. For a glossary of terms and calculations used in Sageworks’ solutions and database, visit our financial terms glossary.
Given the size of the sample used, Sageworks’ Sales Percent Change data typically has a margin of error less than or equal to +/- .77 percent. In other words, if the monthly average in the chart shows an average of 7.00 percent, then we are 90 percent confident that the true average in the population is between 6.23 percent and 7.77 percent. This confidence interval was calculated using the formula:
where x=average of the sample, s=standard deviation of the sample, and n=sample size. Because our sample size is sufficiently large and because the population’s standard deviation is unavailable, we use the sample’s standard deviation. The margin of error for Net Profit Margin is smaller as there is less variability in that sample.
Given the confidential nature of private-company financial information, Sageworks could not collect data from the population of all private companies. Instead, we rely on a sample. We recognize that our sample may have a slightly positive or negative bias because the majority of our data comes from companies who use an external financial advisor, either an accountant or banker. For our calculations, we assume our sample is representative of the population, but we cannot validate that it is perfectly representative.