Industry benchmark · Refreshed October 2026

Every ecommerce AI agent,
tested the same way

The field

Who wins each lane

How every vendor performs on the two axes that decide the category: automation rate and conversational competence. Toggle between pre-sale and post-sale to switch the job, then read the full ranking below, scored by a composite of automation, answer quality, and speed.

Automation rate vs answer quality · pre-sale lane

Up and to the right is better on both axes.

Pre-sale lane
#VendorCompositeAutomationQualityMeanN
Quality by intent

Quality by conversation type

Blind LLM-judge quality (/100) for each conversation type, Gorgias against the ranked field. Baked from judged transcripts in the ranking window.

Shopping Assistant

GorgiasField
Everyday value
76.6
58.6
Gift
77.7
59.8
Budget compare
74.7
55.5
Getting started
72.1
57.1
Problem-solver
67.1
49.4

Gorgias leads every Shopping Assistant intent by 15 to 19.2 points, strongest on budget compare.

Support Agent

GorgiasField
Returns policy
86.9
71.5
Order tracking
75.7
50.8
Shipping policy
77.3
68.3
Damaged item
68.9
59.5
Modify or cancel
60.3
56.5

Gorgias leads every Support Agent intent by 3.8 to 24.9 points, strongest on order tracking.

Distribution

Trends: how each vendor changes over time

One point per week, pooled across that week's captures. A week with fewer than 10 conversations stays blank, so a thin sample can't draw a spike. Gorgias and up to three providers you pick are in color; the rest of the field is grey. Ignores the date window (shows every week).

Distribution

Response speed: where the time goes

Every conversation follows the same arc. T1 is a single direct question. T2 to T7 push discovery and multi-turn reasoning. T8 to T10 force the transactional close: cart, shipping, and policy. Each curve shows where an agent holds up, and where it breaks down.

No cleanly-timed turns for this vendor in the current window.

Shopping-discovery conversations that must end in a real product recommendation.

Window 120 days is the recommended window — it is what the ranking is computed on, and what gives every vendor enough storefronts to be measured fairly.
The rubric

Measure your own AI agent

Download the rubric we used to score every vendor, then grade your own AI agent.

Download the rubric we used to score every vendor, then grade your own AI agent.