R001 — Hardcoded Cells

Purpose¶

R001 detects a hardcoded numeric value embedded in a formula cell, in a row where the surrounding cells otherwise share a clear, dominant formula pattern. It is the FMAE engine's primary check for the single most common structural risk in financial models: a business assumption typed directly into a calculation cell instead of referenced from a controlled input.

Detection Logic¶

R001 examines each row of a worksheet as a group. If a row's interior cells (excluding the first and last cell in the row — a different edge-of-row value, such as a ramp-up first period or a terminal value, is common and legitimate, so the row's edges are excluded from the comparison) establish a clear, dominant formula pattern, any cell within that interior that contains a typed numeric value instead of following the row's formula pattern is flagged. A row that does not establish a clear dominant pattern in the first place — too few comparable cells, or no single formula pattern dominant enough to call a baseline — is not evaluated by this rule at all.

This is a conceptual description of what the rule looks for; the specific thresholds and sample-size requirements behind "a clear, dominant formula pattern" are internal to the engine and not disclosed here.

Why It Matters¶

A hardcoded value inside a formula row does not respond when the model's stated assumptions change elsewhere. If a row's formula is meant to calculate revenue as prior-period revenue times a growth rate held in an input cell, and one period's cell instead contains a typed number, that period's revenue will not move when the growth rate assumption is changed — while every other period in the row correctly does. The model appears to respond to the assumption change (because most of the row does), while silently not applying it to the affected period. This is the specific failure mode that makes hardcoded cells the highest-weighted rule in the entire engine (15.0) and one of five rules capable of capping a model's overall letter grade under the SM-2.0 Scoring Methodology, regardless of how high the model's numeric score otherwise is.

Evidence Produced¶

A triggered R001 finding's evidence identifies the specific sheet and cell address of the hardcoded value, the cell's current value, and a description stating how many of the row's interior cells follow the dominant pattern versus how many (including this one) do not, so the finding is traceable back to exactly the row context that caused it to be flagged. See the Finding Data Model & Confidence page for the full evidence and provenance model this finding is built on.

Severity¶

Critical. Weight 15.0 — the single highest weight of any rule in the FMAE engine. R001 is one of five critical-override rules under SM-2.0 (alongside R002, R004, R006, and R023): if it triggers even once, the model's overall letter grade is capped below A or B regardless of the numeric score the weighted formula produces.

Example¶

Trigger. A revenue row runs five annual formula cells, four of which read =C5*(1+$B$12) (prior period times one plus a growth-rate input cell), shifted appropriately by column. The fifth cell instead reads 92450 — a typed number with no formula at all. R001 flags this fifth cell: the row's interior otherwise establishes a consistent formula pattern, and this cell departs from it with a typed constant.

Non-trigger. The same row, but every interior cell correctly reads a shifted version of =C5*(1+$B$12). No cell departs from the row's pattern, so R001 does not fire — even though the row contains a growth-rate assumption, because that assumption lives in the referenced input cell, not typed into the formula.

Recommended Remediation¶

The typed value is replaced with a formula consistent with the row's dominant pattern, referencing the same input cell the rest of the row already uses. Where the flagged cell's typed value differs from what the input-cell-referenced formula would produce, this should be treated as a separate finding in its own right — the typed value may represent an intentional override that was never reconciled with the model's stated assumption, and simply restoring the formula would silently change the model's output without that change being understood.

Related Rules¶

R001 shares the Structural taxonomy category with R002, R004, and R006, each published in this same initial phase of the Rule Reference; the remaining Structural-category rules follow in later phases and are indexed on the Rule Reference page as they are published. R001 and R004 are, specifically, a general/specific pair: both compare a row's interior cells against its own dominant pattern, but R001 flags a typed constant overriding the pattern while R004 flags a different formula breaking it.

Related Articles¶

• Hardcoded Formulas in Financial Models — the Knowledge Centre's buyer-facing technical guide to hardcoded values, covering how they arise, their structural risk, and manual versus automated detection.

Related Research¶

• RP-001: Why Hardcoded Cells Cause Financial Model Failure — a methodological account of this failure mode, grounded in this rule's detection rationale.

Version History¶

No version-specific changes to R001's severity, weight, or category are recorded as of engine version 2.1.0. This section will be updated in the same edit as any future change to the rule's declared attributes.