✕

How to read the B4 Index

Every category is placed on two axes, then scored on five dimensions.

The four calls

The two axes

Where a category lands comes from two axes, each on a 1–5 scale, split by a line at 3.5:

• Strategic Differentiation (X) = average of Specificity and Strategic Control — how much this is your edge vs. a generic capability.
• AI Feasibility (Y) = the AI Feasibility score — how realistic it is to build or replace this with AI today.

An axis counts as high only above the line — in practice, at 4 or above. BUILD = both high · BRIDGE = X high, Y low · BEWARE = X low, Y high · BUY = both low.

The verdict is banded

The three quadrant-deciding dimensions are read as ranges, not points: one step either side of what each scored, equal weight on all three values. Enumerating all 27 combinations gives a probability for each of the four calls. The one holding the most is the call. A split like BUY 67 · BEWARE 33 reads as: 67% of those readings land on Buy, and 33% land on Beware — the smaller share is how the call could flip.

• Confidence. Clear at 70% or more, lean between 50 and 70, split below 50. A near call flag when the runner-up is within 15 points.
• Ties. Equal weight breaks toward the cheaper mistake: Buy, then Bridge, then Beware, then Build. Build is the last call a tie can produce.
• The lens. The AI-maturity lens in the filter sidebar shifts AI feasibility one step down at low maturity and one step up at high. It changes how each verdict reads, never which categories match, and nothing is stored.

Methodology v4.0 — August 2026. Full framework →

The five scores (1–5)

Specificity.

How specific the capability is to how your company operates, rather than a generic pattern. → feeds Strategic Differentiation.

Strategic Control.

How much owning this matters to your control and differentiation. → feeds Strategic Differentiation.

AI Feasibility.

How feasible it is to build or replace this with AI now. → the Y axis.

Vendor Value.

How much commercial vendors genuinely add (R&D, scale, compliance). → feeds Urgency.

Cost Trajectory.

Where vendor pricing is heading. → feeds Urgency.

Urgency = average of Vendor Value and Cost Trajectory (High ≥3.5, Medium 2.5–3.4, Low <2.5).

Vendors

Vendors are first-class in the index — 23,000+ vendor-to-category mappings sit under the scores.

• In the directory. Expand any category row for its vendors and observed pricing; the vendor count is a sortable column.
• Vendor pages. Every qualified vendor has a fact page at /vendors/<name>: what it is, the categories it serves, funding history, and when the index last verified it alive. The category verdict on those pages is Pro-only, same as everywhere else.

The Continuum

Categories move, and the index logs every move. The Continuum tab at the top of this app holds the full movement history: every verdict change the scoring engine has recorded, newest first, filterable by landing verdict and domain, with the reasoning expandable on each row. Whole-index methodology passes are tagged so organic market movement stays distinct.

The public /continuum page shows the 20 most recent moves and the engine's status; this tab is the complete record.

How the index is organized

Function group.

The broadest grouping — 20 business functions (Engineering/IT/AI, Sales/Marketing/CX, Operations/Supply Chain, and so on). Each holds several domains.

Domain.

A shelf of related software categories inside a function group — e.g. Marketing Technology sits under Sales, Marketing & CX. 169 in all.

Category.

The scored atom — a specific software need like Configure Price Quote (CPQ). Each gets its own verdict and scorecard.

Industry.

Which industries a category serves. Most are Cross-industry (they apply everywhere); the rest are vertical tags like Mining or Live Events. A category can carry several.