METHODOLOGY

How we measure your e-shop's AI readiness

The AgentReady audit evaluates an e-shop's readiness for AI recommendations across three equally weighted dimensions. Each scores 0–100 and comes with a prioritized action plan and a benchmark against your competitors.

CORE PRINCIPLE

"AI doesn't recommend the cheapest product — it recommends the best described one."

AI prioritizes well-described, trustworthy products that match the customer's intent — not the cheaper variant lacking context.

THREE EVALUATION PILLARS

Each pillar 0–100 points · three independent scores

PILLAR 10 — 100 points

Product Content Score

The quality and completeness of every product's data. How detailed, structured and machine-readable your catalog is.

Example

A handbag retailer jumped from 35/100 to 72/100 after filling in missing materials, dimensions and use-case data.

What we evaluate

Parameter completeness — tech specs, dimensions, material, compatibility
Description depth — use cases, scenarios, benefits vs. alternatives
Data structure — attributes, categorization, variants ready for filtering
Catalog consistency — the same quality across all product types
Machine readability — JSON-LD, structured formats, clean feeds
AI enrichment — usage scenarios and customer search intent

PILLAR 20 — 100 points

Entity Trust Score

The overall trustworthiness of the e-shop. How platforms, reviews and customers perceive you beyond your own site.

Example

An electronics shop had perfect product data, but only a 40/100 trust score because of missing Heureka reviews, 5-day delivery times and a missing FAQ.

What we evaluate

Reputation and ratings — product reviews and shop ratings on external platforms
Operational reliability — SLA performance, stock availability, delivery times, returns
Customer support — FAQs, guides, quality of documentation and replies
Brand consistency — a unified presence across marketplaces and channels
Business transparency — clear terms, contact details, company ID, address
Reputation signals — time in business, review volume, rating distribution

PILLAR 30 — 100 points

Recommendation Fit Score

How well your data matches real customer questions and when you actually show up in AI recommendations.

Example

A swimwear retailer had no size tables or body-type information — AI didn't recommend it for specific needs ("swimwear for curvy figures").

What we evaluate

Coverage of real queries — does your data answer "best swimwear for curvy figures"
Decision scenarios — content supports selection, comparison and recommendation
AI visibility — how often you appear in relevant AI answers
Competitor awareness — differentiation versus rivals in the category
Contextual relevance — alignment with seasonality and current trends
Reviews and reputation — external endorsements that back up your product claims

READINESS LEVELS

How to interpret the overall score

ScoreLevelCatalog status

80 — 100ExcellentYour catalog is fully ready for AI recommendations across all channels.

60 — 79GoodA solid foundation with minor gaps. AI picks you sometimes, but not consistently.

40 — 59Needs workSignificant gaps. AI will likely skip you in favor of competitors.

0 — 39CriticalInsufficient data for any recommendation. You're not visible in AI shortlists at all.

WHAT YOU GET

Concrete audit deliverables

3 separate scores

Product Content, Entity Trust and Recommendation Fit — plus an averaged overall score.

Prioritized action plan

Ordered by impact: what to fix first, what comes next, what can wait.

Competitive benchmarking

How you stack up against the top 3 competitors in your category.

Audit your e-shop in 2 minutes

Free analysis with no signup — you get 3 scores, a prioritized action plan and a competitor comparison.

Start the audit for free