If you’ve used the original BESHStat (VBA / .xlam) add-in over the years, you already know the idea: keep statistics close to the spreadsheet and make analysis fast and practical.
Now I’m rebuilding it as BESHStatNG (New Generation) — a VB.NET + Excel-DNA add-in distributed as a lightweight .xll.
Why re-implement BESHStat from VBA to VB.NET?
VBA served its purpose well, but it comes with limits that become painful as a project grows. Moving to VB.NET brings a lot of practical advantages:
Performance and stability
Compiled .NET code runs faster and is easier to keep stable as features expand.Better structure = easier maintenance
Clean separation of UI, computations, and output formatting means fewer regressions and faster development.More reliable numerical routines
BESHStatNG intentionally avoids Excel worksheet functions for core statistics and uses internal math routines designed for tail stability and reproducibility.No heavy external dependencies
The add-in is intentionally kept small (currently ~1.5 MB) while still providing a broad range of statistical tools.A modern foundation for future features
Better UI patterns, more complex models, and richer workflows are much easier to build in .NET than in VBA.
Current progress
The VB.NET version is now roughly ~95% of the original VBA functionality re-implemented, with the focus on:
Assumptions and diagnostics (normality, outliers, variance homogeneity, symmetry)
Core descriptive statistics
Visualization tools (including custom charts that work even on older Excel versions)
A consistent “Excel-first” workflow: select data → choose options → generate outputs into a new worksheet
What you can look forward to
Here are examples of what’s already working (and what you’ll see expanding):
Normality tests + diagnostics
quick test results + interpretation-friendly plots
Outlier detection
single outlier (Grubbs) and multiple outliers (Rosner/ESD)
Homogeneity of variance
multiple tests available (robust + normal-theory options)
Symmetry testing and symmetry plots
fast distribution diagnostics beyond “just normality”
Box-and-whiskers plots
generated by the add-in (this feature was built before Excel offered modern boxplots by default)
New: online help & documentation (work in progress)
To make the add-in usable without needing any source code access, I’m building a dedicated multi-page HTML help site (generated with MkDocs and hosted on my website). It covers:
Getting started (install + first steps)
How to select data (range picker vs multi-variable selection)
Each statistical method with screenshots, options, and interpretation guidance
👉 Help site: https://www.beshstat.eu/beshstatng/help/latest/
Want to test it?
The installer is not code-signed, so Windows may show a “proceed anyway” warning during installation. If you’re comfortable with that and want to try the new generation add-in, keep an eye on this page for updates.
Follow along
I’ll post updates as I finish remaining modules, polish the UI, and expand the documentation with more method pages and examples.
If you’ve used the old VBA add-in and have a favorite feature or workflow you want prioritized in BESHStatNG, send me a message — real user feedback helps shape what comes next.
— BESH