Compensation Studies

It starts from what the shareholder actually does

Describe the role the way you would to a colleague and it comes back with three to six duties, each matched to an occupation code. You correct them — that is the point of the step. Someone who spends a third of the year selling and two thirds preparing returns is two duties at two different wage bases, and no single job title says that.

• The model can only choose codes that have published wage data behind them
• Every duty carries your share-of-time split and the percentile you picked
• Tax year, geography, annual hours and years of experience are all inputs

The number is arithmetic, not an opinion

Each duty resolves against the Bureau of Labor Statistics wage survey — the same data a valuation analyst reaches for — at the percentile you chose, for the year and area on the study. The composite is those blended by how much of the year each duty takes. The AI never proposes a salary. It proposes duties, and duties resolve to published wages.

• Wages by SOC code and geography, at the p10 / p25 / p50 / p75 / p90 percentile you pick
• Composite computed from the duty mix rather than from one job title
• The survey year is recorded, so a memo says which vintage it was built on
• Signing refuses outright if a duty has no published wage for that code and area

And it asks the question an examiner will

The cost approach gives you a figure. The independent investor test asks whether what is left after paying it still earns a fair return on the shareholder’s equity. It computes the residual and shows it against the return you require, as a pass or fail sitting next to the number — rather than as an exercise everyone knows about and nobody does.

The memo is the deliverable, and signing freezes it

A sectioned memo on your letterhead — background, method, the investor cross-check, the factors considered, the conclusion — rendered to PDF and filed on the client as a workpaper for that year. Signing locks every figure in it, so a later revision to the wage data cannot quietly restate a memo you already handed someone.

• Section IV sets out the multi-factor test the courts apply, not just the arithmetic
• Each duty line cites its SOC code and the percentile used
• The data source and survey year are stated in the footer
• A signed study cannot be edited — corrections go through an amendment that links to the original
• The shareholder can be sent the memo to acknowledge, if your firm works that way

No plan decides which clients get one

The specialist tools meter this: ten clients and one user on the entry plan, fifty and three on the next. That cap ends up choosing which of your S-corps get looked at — the obvious ones, and not the marginal one whose number has drifted for three years. Here it is on the seat, for every client, for everyone.