Free marketing tools Experiments and statistics
Calculators for deciding whether a change worked. Each one states its method, so you can check it against your own tools.
Conversion rates, A/B significance and the statistics behind them.
Conversion rate calculator
Visitors and conversions in, rate out — plus an A/B comparison with a two-proportion z-test.
- 95% confidence interval for each group
- Absolute and relative lift
- Plain-language significance verdict
Percentage change calculator
Percentage increase or decrease between two values, the absolute difference and the new value as a share of the old.
- Works with negative values
- Explains percent vs percentage points
- Flags a zero starting value
Growth rate calculator
Compound growth per period between a start and an end value, plus total growth and the simple average.
- Per-period compound rate
- Total growth over the whole span
- Shows why the simple average overstates
CAGR calculator
Compound annual growth rate from a starting value, an ending value and the number of years, with the total multiple and the years to double.
- Accepts partial years such as 2.5
- Total multiple and doubling time
- Formula and worked example on the page
A/B test sample size calculator
Visitors needed per variant to detect a given lift in conversion rate, and how many days your traffic needs to get there.
- Baseline rate and relative minimum detectable effect
- Significance level and power options
- Test duration from daily traffic
P-value calculator
Turn a z-score or t-statistic into a p-value, one- or two-tailed, with the critical value at the 5% level.
- z-test or t-test with degrees of freedom
- One- or two-tailed
- Critical value and a plain verdict
Confidence interval calculator
The range a conversion rate or an average probably falls in, from your sample, at 90%, 95% or 99% confidence.
- Wilson interval for proportions
- Student t interval for means
- Margin of error and both bounds
A/B test duration calculator
How many days an A/B test needs from your traffic, baseline rate and the lift to detect, rounded to whole weeks.
- Relative or absolute minimum detectable effect
- Traffic share and number of variants
- Whole-week plan and long-test warning
Minimum detectable effect calculator
The smallest lift an A/B test can reliably detect with the traffic and time you have, with a table by test length.
- From traffic × days or a fixed sample
- Relative and absolute MDE
- MDE for 1 to 8 weeks
SRM checker
Check whether an A/B test’s groups got the traffic split you configured, with a chi-square test and the usual SRM threshold.
- 2 to 6 groups, any expected split
- Chi-square, p-value and per-group deviation
- Likely causes when SRM is found
Bayesian A/B test calculator
Probability that B beats A, the expected loss of each choice and credible intervals, from visitors and conversions.
- Exact numerical integration, not random sampling
- Uniform, Jeffreys or custom prior
- p-value shown alongside for comparison
T-test calculator
Compare two means from raw data or from mean, SD and n: Welch, Student, paired or one-sample, with p-value, confidence interval and Cohen’s d.
- Welch, Student, paired, one-sample
- Paste raw data or enter summary stats
- CI of the difference and Cohen’s d
Chi-square calculator
Test a contingency table for independence or observed counts against expected ratios. Get χ², degrees of freedom, p-value, expected counts and Cramér’s V.
- Up to 6 × 6 tables, paste from a spreadsheet
- Goodness-of-fit mode
- Yates correction and Fisher’s exact test for 2 × 2
Standard deviation calculator
Paste a list of numbers to get sample and population standard deviation, variance, mean, median, quartiles and the formula worked through with your data.
- Sample (n − 1) and population (n)
- Quartiles with the method named
- Every formula step with your numbers
Survey sample size calculator
How many responses a survey needs for a given margin of error and confidence level, with finite population correction and the number of invitations to send.
- Margin of error, confidence, proportion
- Finite population correction
- Invitations from expected response rate
Margin of error calculator
The ± margin of a survey percentage from sample size, proportion and confidence level, with optional population correction and a table by sample size.
- Any confidence level, any proportion
- Finite population correction
- Margin of error table by n
Weighted average calculator
Pair each value with its weight, or paste two columns from a spreadsheet, and see the weighted average next to the simple average, with Σ(x·w) ÷ Σw worked through.
- Weights as percentages or any numbers
- Paste labels, values and weights together
- Each item’s contribution to the result
Z-score calculator
Turn a value into a z-score, percentile and p-value, find the value at a given percentile, or paste a column to flag outliers with z and modified z.
- z, percentile, one- and two-tailed p
- Percentile → z → value
- Z-score of every value in a list
Linear regression calculator
Paste two columns to get the regression line, Pearson r, R², standard errors, t and p, predictions with intervals, Spearman’s rho and a scatter plot.
- Slope, intercept, r, R², p-value
- Prediction and confidence intervals
- Scatter plot with the fitted line
Questions about the tools
Still have a question?
Chat with the team behind TapCub — we usually reply within a few hours on business days.
Do I need a TapCub account to use them?
No. The tools are public pages. An account only matters when you want the same numbers computed from your live traffic instead of typed in.
Is anything I type sent to TapCub?
No. Almost every tool runs entirely in your browser. A few lookup tools (DNS, WHOIS, page speed) send only the domain or URL you enter, from your browser straight to a public service such as Cloudflare DNS, the domain registry or Google PageSpeed, and only when you press the button; each of those pages says where the request goes. We record that a tool page was viewed and used, like any other page — never the values.
Are the tools kept up to date?
Yes. Every tool page lists its version, an update history and the date each data source was last checked. Data such as crawler rules, public holidays and platform limits is re-checked on a schedule, and all tool updates are listed in one place.
Can I embed or link to a tool?
Linking is welcome. The pages have stable URLs under /tools. We do not offer an embed, but the formulas are documented on each page so you can reproduce them.
Why do my numbers differ from my analytics tool?
Usually the definition, not the arithmetic: who counts as a visitor, which event converts, and whether bots were removed first. Each tool lists the definition it assumes. See why UV never matches.
Will there be more tools?
When a question comes up often in support we turn it into a page. Suggestions go to contact.