In Practice

In Practice is a library of working reference builds, studies and notes on how problems are approached.

Each entry includes the workbook and its accompanying documentation. There is no login, no form and no fee.

The public workbooks are interactive but locked. You can change the intended inputs, explore the outputs and see how the logic behaves. They demonstrate the method without transferring the underlying build.

The documentation goes further. It explains the problem, the choices made, the logic, the limitations and where the approach may be useful. It is written for the person evaluating the work, not only for the person using it.

Some entries are practical builds. Others examine a method, a technical question or a broader operational problem. Together, they make the delivery standard visible: how I frame a problem, structure the reasoning and turn it into something practical.

These are reference builds, not ready-made solutions. Bespoke work begins where the examples end: with your files, users, rules and decisions.

Below is the current library of reference builds. Each entry explains the work, what is included and how to access the files. Where available, a short video is also included to introduce the build and show how the solution is configured and used in practice.

A positional method for partitioning a finite numerical dataset into intervals defined by user-supplied boundaries. Observations and boundaries are embedded within a single ordered array and the count for each interval is recovered directly from the positions of the boundaries rather than from separate conditional logic.

The value is in what LET and LAMBDA make possible. Each stage of the construction, the embedded array, the boundary positions, the interval counts, is named and kept visible rather than compressed into an opaque formula. The result is a single reusable worksheet function, PHR_PARTITION, that returns the lower endpoint, upper endpoint and count for every interval. The same approach extends to any boundary driven task, from pricing tiers to risk grades, without rebuilding the logic each time.

Included: White Paper PDF.

Access: Via this link.

A technical case study and Excel workbook built around Wikipedia’s worked logistic regression example. Python in Excel checks the fitted coefficients. Native Excel then rebuilds the same result using LET functions, row level log likelihood calculations and Solver.

The result is not treated as a black box. The workbook shows the route from data to probability, likelihood and fitted coefficients.

Included: Case study PDF and Excel workbook.

Access: Via this link.

These files are hosted in the public Phronesis shared folder. Excel workbooks should be downloaded and opened in Microsoft Excel.
Google Sheets preview may show errors, as Python in Excel, Solver, dynamic formulas and interactive workbook elements are not supported there.

An Excel workbook built around Eurostat’s monthly unemployment dataset. Power Query transforms the raw source into structured tables. A dynamic interface then lets users select countries, population groups and time periods, with results shown through a chart and a country level summary table.

The data arrives in its published form. This solution makes it ready to work with.

Included: Workbook, Eurostat source data and user guide.

Access: Via this link.

These files are hosted in the public Phronesis shared folder. Excel workbooks should be downloaded and opened in Microsoft Excel.
Google Sheets preview may show errors, as some formulas, macros and interactive elements are not supported there.

An Excel workbook built around a budget allocation problem. VBA generates all feasible whole number combinations under a budget cap and minimum unit requirement. A LET formula then returns the closest valid option within the selected budget, with the full option space available for review in a separate table.

The problem looks simple until the combinations grow. This solution makes the feasible choices visible.

Included: Workbook and user guide.

Access: Via this link.

Important: If Excel shows a pink security warning saying that macros have been blocked because the file source is untrusted, please refer to the Workbook Info tab for guidance on how to proceed.

These files are hosted in the public Phronesis shared folder. Excel workbooks should be downloaded and opened in Microsoft Excel.
Google Sheets preview may show errors, as some formulas, macros and interactive elements are not supported there.

An Excel workbook that searches across multiple tables at once and identifies where selected keywords appear. It gives users a controlled first filter before the manual review begins.

Most of the work is in knowing where to look. This solution does that part.

Included: Workbook, sample data and user guide.

Access: Via this link.

These files are hosted in the public Phronesis shared folder. Excel workbooks should be downloaded and opened in Microsoft Excel.
Google Sheets preview may show errors, as some formulas, queries and interactive elements are not supported there.

An Excel workbook that examines π in two ways. Power Query searches for a selected 6-digit sequence inside a structured table containing the first 1 million decimal digits of π. VBA then runs a live Monte Carlo simulation, generating 2,000 random points to show how π can be approximated through sampling.

This solution shows how Excel can search large ordered datasets and run controlled simulations inside a workbook teams already know.

Included: Workbook, source data and user guide.

Access: Via this link.

Important: If Excel shows a pink security warning saying that macros have been blocked because the file source is untrusted, please refer to the Workbook Info tab for guidance on how to proceed.

These files are hosted in the public Phronesis shared folder. Excel workbooks should be downloaded and opened in Microsoft Excel.
Google Sheets preview may show errors, as some formulas, queries, macros and interactive elements are not supported there.