How the matching model works

The ranking is rule-based. Affiliate status is not a ranking input.

Inputs

The finder currently uses six user-selected signals: project purpose, skill level, primary priority, monthly budget band, ecommerce requirement and whether the visitor prefers AI generation or AI-assisted manual control.

Product data

Each product is represented by structured tags for documented capabilities, intended user level and broad budget fit. We use vendor product pages as the primary source for capability checks. A product earns points only where its tags match the visitor's answers.

Weighting

Purpose and ecommerce compatibility receive the highest weight because a mismatch there can make a product unsuitable regardless of design preference. Priority, experience, budget and AI workflow are secondary fit signals.

Commercial independence

Affiliate links may be added after partner approval. Whether a vendor pays commission does not change the rules used to score it. The commercial model is referral-based: AI BuilderMatch may earn a commission if a visitor follows an eligible link and buys.

MVP limitation

This version is intentionally small. If the asset earns real search traffic, the next stage is hands-on benchmark testing, pricing refreshes, more granular business-use cases and change monitoring.

Last updated: 11 October 2026.