The Instafill Form Complexity Index

A transparent 0–100 score that estimates how much effort a form demands to complete — the higher the score, the harder the form. Every score is computed deterministically from the form’s own structure using the formula below, so anyone can reproduce it. Nothing is hand-assigned.

Across the Instafill catalog

Instafill has scored 3,468 forms with the Form Complexity Index. The median score is 50 / 100. Here is how the catalog distributes across the five tiers:

TierScore rangeFormsShare
Simple 0–28 204 6%
Basic 29–45 1,251 36%
Moderate 46–62 991 29%
Complex 63–79 782 23%
Very Complex 80–100 240 7%

The formula

The index has two parts. A base — a weighted sum of five components, each scaled to 0–100 — captures the raw work a form imposes. Three modifiers are then added on top to reward the things that make a form disproportionately hard but that the base can miss: tables, bundled instructions, and dense page layouts. Each modifier is near zero for an ordinary form and only lifts genuinely demanding ones. The total is clamped to 0–100 and rounded:

FCI = clamp(  base  +  T  +  I  +  Y , 0, 100 )
where  base = 0.36·F  +  0.26·D  +  0.15·L  +  0.16·C  +  0.07·S

Base components

ComponentWeightDefinition
F — Field load0.36norm(fields, 3, 200) — the count of fillable fields, log-scaled.
D — Input difficulty0.26100 × (Σ weight(field type)) / fields — the average per-field difficulty (table below).
L — Length0.15clamp(100 × (pages − 1) / 11) — fillable pages; 12+ pages reaches 100.
C — Conditional logic0.16clamp(100 × dependencies / fields) — share of fields gated by “fill only if…” rules.
S — Structure0.07norm(sections, 1, 30) — the number of distinct field groups, log-scaled.

Complexity modifiers

Each modifier adds points on top of the base — up to the cap shown — and contributes nothing to a form that lacks the feature, so a simple form scores exactly as its base.

ModifierMaxDefinition
T — Tables & lists+13norm(tables, 0, 15) × 0.13 — from the number of detected tables and repeating-row lists. Grids of line items take real effort the field count alone understates.
I — Instructions+9clamp(100 × instruction‑pages / 12) × 0.10 — pages of printed instructions (total pages − fillable pages). A form with few fields but many pages of rules to read is still hard.
Y — Layout density+12clamp((norm(density, 15, 120) − 34) × 0.30, 0, 12), where density = (fields + groups) / fillable pages — only crowded, information-dense pages clear the floor.

The log-scaling function

Field and section counts are compressed on a log scale so that the difference between a 5- and a 50-field form matters more than the difference between a 450- and a 500-field form:

norm(x, min, max) = clamp( 100 × [ ln(x+1) − ln(min+1) ] / [ ln(max+1) − ln(min+1) ] , 0, 100 )

Input-difficulty weights

Each field contributes a difficulty weight based on what the user must do to complete it:

Field typeWeightField typeWeight
Signature0.9Dropdown / Combo box0.4
Free text0.6Radio button0.4
Number / Date / Time0.5Checkbox0.2
Any other type0.5Button0.2

Complexity tiers

TierScoreTypical form
Simple0–28A short, single-page form with a handful of fields.
Basic29–45A common one- or two-page form (e.g. IRS Form W-9).
Moderate46–62A multi-page form with real work (e.g. SSA-44).
Complex63–79A long or heavily conditional form (e.g. Form 5695, Form 8949).
Very Complex80–100An exceptionally long, table-heavy, or instruction-bundled form (e.g. the 603-field ACORD 125).

Worked examples

All three examples below are the live scores for these forms — you can verify them against each form’s page. They show the base, then each modifier, then the final index.

Form W-9  Basic · 39

23 fillable fields (15 free text, 8 checkboxes) · 1 fillable page · 5 instruction pages · 8 groups · 4 conditional fields · no tables

ComponentCalculationValue
Fnorm(23, 3, 200)46
D100 × (15×0.6 + 8×0.2) / 23 = 100 × 10.6/2346
L100 × (1 − 1) / 110
C100 × 4 / 2317
Snorm(8, 1, 30)55
base0.36×46 + 0.26×46 + 0.15×0 + 0.16×17 + 0.07×5535
+ Iclamp(100 × 5 / 12) × 0.10  — 5 instruction pages+4
FCI35 + 4  (T and Y add 0)39

Form SSA-44  Moderate · 54

54 fillable fields (25 free text, 20 checkboxes, 7 numbers, 1 date, 1 button) · 8 fillable pages · 24 groups · 9 conditional fields · no tables or instruction pages

ComponentCalculationValue
Fnorm(54, 3, 200)67
D100 × (25×0.6 + 20×0.2 + 7×0.5 + 1×0.5 + 1×0.2) / 54 = 100 × 23.2/5443
L100 × (8 − 1) / 1164
C100 × 9 / 5417
Snorm(24, 1, 30)92
base0.36×67 + 0.26×43 + 0.15×64 + 0.16×17 + 0.07×9254
FCI54  (no modifiers apply → FCI = base)54

ACORD 125 — Commercial Insurance Application  Very Complex · 81

603 fillable fields · 4 fillable pages (≈151 fields/page) · 44 groups · 48 conditional fields · 5 tables

ComponentCalculationValue
Fnorm(603, 3, 200)  (maxes out)100
D100 × (Σ difficulty) / 60349
L100 × (4 − 1) / 1127
C100 × 48 / 6038
Snorm(44, 1, 30)  (maxes out)100
base0.36×100 + 0.26×49 + 0.15×27 + 0.16×8 + 0.07×10061
+ Tnorm(5, 0, 15) × 0.13  — 5 tables+8
+ Yclamp((norm(162, 15, 120) − 34) × 0.30, 0, 12)  — density (fields+groups)/page ≈ 162+12
FCI61 + 8 + 12  (I adds 0)81

What is counted

Field counts include only active, fillable fields — static labels and fields excluded from autofill are not counted. The Length component counts only pages that carry fields, so a form is never penalized as “long” for printed instructions it bundles; those instruction pages are measured separately by the Instructions modifier. Tables and repeating lists are detected from the form’s own layout, and density is fields-plus-groups per fillable page. “Conditional fields” are fields whose value depends on an answer earlier in the form.

How the methodology evolves

The Form Complexity Index is a living standard, not a one-time guess. Every version is published with what changed and why, and each stored score records the version that produced it — so a number is always traceable to an exact, reproducible formula. We recalibrate as we learn from the full catalog.

v2.0 Tables, instructions & layout density — current

Scoring the entire catalog with v1.0 exposed a blind spot: the base rewards depth — more fields, more pages, more branching — but treated a dense grid of line items, a thick instruction booklet, and a roomy one-pager as roughly equal, so almost no form reached Very Complex. v2.0 keeps the v1.0 base untouched and adds three additive modifiers — Tables, Instructions, and layout Density — that lift only the forms that are hard along an axis the base can’t see. We then recalibrated the tier cutoffs against the live distribution, so every tier from Simple to Very Complex is now meaningfully populated.

v1.0 The five-factor base — foundational

The first published index established the core model: a weighted blend of field load, input difficulty, page length, conditional logic, and structure, each normalized to 0–100. The weights weren’t chosen in the abstract — we calibrated them against dozens of well-known forms (W-9, SSA-44, Form 5695 and others) and tuned until the resulting tiers matched the judgment of people who fill these forms for a living.

Methodology v2.0. The Form Complexity Index is published in full so it can be referenced and independently verified. Each form’s score, component breakdown, and standardized identifiers — including the federal OMB control number — appear in the “Form Complexity” panel on that form’s page.