Smart Data Profiler

Q: Is the Smart Profiler free?

Yes. You can run it without an account (3 free analyses), and a free account adds 10 more — a one-time allowance, no payment needed. Every result includes the APA 7 write-up. Paid plans (Student from $9/month) add more analyses and Word, PDF and Excel export.

Q: How does it detect outliers?

Outliers are flagged using the IQR fence method (values below Q1 − 1.5×IQR or above Q3 + 1.5×IQR). This is robust to non-normal distributions.

Q: What normality test does it use?

Shapiro-Wilk for columns with fewer than 50 observations and the Lilliefors (Kolmogorov-Smirnov) test for larger samples.

Q: What should I check before running statistics on a new dataset?

Check that each column has the right type (numeric vs categorical), look for impossible values (e.g., ages of 999 used as missing codes), count missing values per variable, inspect distributions for skew and outliers, and confirm that IDs are excluded from analyses.

Q: Does profiling change my data?

No. A profile only reads your data and reports on it; recoding or excluding anything is a separate step that you request explicitly.

One-click intelligent profiling — instantly understand the shape, quality, and quirks of any dataset.

The Smart Data Profiler analyses every column in your dataset and returns a structured quality and distribution report in seconds. It classifies variables as numeric, categorical, or datetime; measures skewness and kurtosis; runs Shapiro-Wilk or Lilliefors normality checks; counts outliers via the IQR fence method; and flags columns with high missing-data rates.

Understanding your data before running statistical tests is essential for correct analysis selection and valid results. The profiler identifies which columns might violate normality assumptions, which have extreme skew that warrants transformation, and which have enough unique values to be treated as continuous versus ordinal. This guides the entire analytical workflow.

StatRyx's profiler also produces a visualisation panel with histograms for numeric columns and bar charts for categorical columns, giving you an immediate visual sense of the data landscape. A concise quality summary flags any issues requiring attention before analysis.

Profiling a dataset before analysis catches the problems that silently break results later: numbers stored as text, Likert answers typed as words, ID columns treated as variables, a few extreme values, or one variable missing for a quarter of the sample. A profile lists each column's type, missing count, range and distribution shape, so you can decide what to recode, exclude or check before choosing a test.

When to use the Smart Data Profiler

You've just imported a new dataset and want a quick overview of its structure and quality
You need to decide which statistical tests are appropriate for your variables
You want to identify columns with too many missing values or extreme outliers before analysis
You're preparing a data quality report for a research team or client

When not to use it

You need formal hypothesis testing on distributions — use Normality Test instead
You need detailed outlier analysis with removal options — use Outlier Detection instead
You need a full missing-data imputation workflow — use Missing Data Analysis instead

Worked example

Variable	Type	Missing %	Skewness	Outliers	Normal?
age	numeric	0.0%	0.42	2	Yes
income	numeric	3.2%	2.81	14	No
satisfaction	numeric	0.0%	−0.19	0	Yes
region	categorical	1.1%	—	—	—

How to report it (APA 7)

Preliminary data screening identified three variables with skewness exceeding |2.0| and one variable with 18.4% missing data. Two variables showed significant departures from normality (Shapiro-Wilk W < .95, p < .05), warranting the use of nonparametric tests.

Frequently asked questions

Is the Smart Profiler free?