Thirty focused days will not make you a statistician. It will comfortably make you someone who can analyse a business dataset competently, read published analysis critically and hold their own in a meeting about numbers. Here is the plan, assuming around an hour a day.
Days 1-5: describing data
Mean, median, mode, range, variance, standard deviation, percentiles. Learn each by hand once on ten numbers, then in a spreadsheet on a real dataset. Milestone: you can explain why the mean and median of the same data differ, and which to use when.
Days 6-8: charts and distributions
Histograms, box plots, line charts, scatter plots. Learn the shape of a normal distribution and what "skewed" looks like. Milestone: you can look at a histogram and say something useful about the data without computing anything.
Days 9-12: sampling and bias
Populations versus samples, random sampling, selection bias, non-response bias, survey design. Spend a full day just on who is missing from datasets - it is the most valuable idea in the entire month. Milestone: you can list three ways a customer survey could mislead you.
Days 13-16: probability basics
Simple probability rules, independence, conditional probability, expected value. Keep it light; you need enough to understand inference, not a formal course. Milestone: you understand why a 95% confidence level does not mean "95% chance this is right".
Days 17-22: inference
Standard error, confidence intervals, hypothesis testing, p-values, statistical significance. This is the hardest stretch - go slowly and do lots of small examples. Milestone: you can state a business result as an interval with a confidence level.
Days 23-26: relationships
Correlation, simple linear regression, interpreting coefficients and R-squared, and the causation caveat. Milestone: you can build a regression in a spreadsheet and explain honestly what it does and does not show.
Days 27-28: forecasting
Trend, seasonality, moving averages, simple projection with an uncertainty range. Milestone: a defensible forecast for one of your own metrics.
Days 29-30: the project
Take one real dataset from your own work and produce a single page: the question, the method, the finding, the confidence range and the caveats. This is what makes the month stick, and it is the artefact you can show at an interview.
What to deliberately skip
Hand-computing anything beyond the first week, mathematical derivations, advanced test variants, and the temptation to learn Python before you understand the statistics. Depth without foundations is wasted effort.
What 30 days cannot give you
Feedback on your reasoning, which is where self-study consistently falls short. You can learn the arithmetic alone; you cannot easily discover that your interpretation of a p-value is subtly wrong. That is the main argument for structured teaching afterwards.
If you want the 30 days to be followed by properly assessed, applied work, the Statistics for Business course picks up from here. Enquire today.