How Alternatives Are Computed

Overview

Every tool on AIDIRECTORY has an alternatives section showing similar tools you might want to consider. We compute these automatically using a scoring formula based on Jaccard similarity - measuring what proportion of tags two tools share, not just how many they have in common.

How the Scoring Works

When we compute alternatives for a tool, we look at every other published tool that passes the hard gates (see below) and calculate a relevance score. The higher the score, the more relevant the alternative.

Jaccard Similarity (Tag Scoring)

Instead of awarding fixed points per matching tag, we use Jaccard similarity for each tag group:

Jaccard = shared tags / union of tags

This measures the proportion of combined attributes two tools share. A tool with 20 tags that shares 3 with another tool scores much lower than a tool with 5 tags that shares 3 - because proportional overlap is a stronger signal of true similarity.

Only Feature and Use Case tag groups contribute to the score. Other groups (Audience, Pricing, etc.) are excluded because they don't indicate whether two tools are functionally interchangeable.

Feature and Use Case are weighted equally. Feature tags capture what a tool can do, while Use Case tags capture how users apply it — both are strong signals that two tools are genuine alternatives.

Hard Requirements

A tool must pass two gates before it can be considered as an alternative:

1. Shared category - must share at least 1 category with the source tool. Tools in completely different categories are never scored.
2. Shared Feature tag - must share at least 1 tag in the "Feature" group. This prevents tools from qualifying as alternatives based solely on pricing model or category match without any actual feature overlap.

Minimum Score Threshold

Any candidate with a score below 7 is dropped. This keeps our alternatives list genuinely relevant - no noise, no random matches.

When Alternatives Are Computed

We compute alternatives at several points:

• When a tool is published - alternatives are calculated automatically.
• On request - there's an option to regenerate alternatives when needed.
• Cascade recomputation - when a tool's categories or tags change, we recompute alternatives for related tools so everything stays up to date.

How Alternatives Are Displayed

All alternatives are pre-computed and stored with a relevance score. Every surface - tool cards, the tool detail page, and the full alternatives page - reads from this same pre-computed data. This ensures consistent, scored results everywhere.

Note: Verified status does not affect the alternatives score. Alternatives are based purely on tool similarity - what the tools do and how they're used.