New: Use Eli in ChatGPT and Claude

Free tool · No signup

Which sources does AI rely on in your market?

Enter a market. We run a representative set of buyer questions, collect every cited source, and aggregate them by domain, page and source type.

Live AI checks open soon

This tool sends real buyer questions to AI systems. We are verifying it against live providers before opening it, so it never shows a result that was not actually measured. These checks are available now:

What it checks

The top influence sources and where you need external authority.

01

Derived from real answers

Every source in the map was returned by a provider for a tested prompt. Nothing is a pre-written example.

02

Four kinds of opportunity

Owned, earned authority, community and partnership, so each source maps to a team that can act on it.

03

Brands referenced

For each source, the brands recommended in the answers that cited it.

When to use it

Example use cases

  • Entering a new market: learn which third parties AI treats as authorities there.
  • Planning PR and review programmes around sources that actually appear in answers.
  • Checking whether a marketplace or community matters before investing in it.

Methodology

How the result is produced

Analysis version 1.0.0 · Scoring methodology 2026-10-04.1 · Updated 2026-10-04

  1. Buyer questions are generated from fixed, published templates using your category, audience and market. Informational prompts such as “what is X” are excluded on purpose.
  2. Each tested question is sent once, unchanged, to every AI provider that is configured, with live web search enabled. Providers that are not configured are listed as “Not tested”. Nothing is simulated.
  3. Sources are aggregated by domain and page with the number of appearances and the prompt families that produced them.
  4. Source types are assigned by fixed rules. Opportunity groups are derived from the type: owned (your domain), earned authority (review, media, publisher, comparison), community, partnership (marketplaces and integrations).
  5. If a provider returns no citations, the map says so. It is never filled with typical examples.

If something cannot be tested, the report says “Not tested” or “Unable to verify”. Uncertainty is never turned into a score.

FAQ

Questions about this tool

Which AI systems are tested?

The report lists every provider with its status. A provider is tested only when its API is configured and answered. Anything else is labelled “Not tested”, never estimated.

Will I get the same result if I run it again?

Not necessarily. AI answers vary between runs, users and locations. This is one observed sample, cached for a week so the report stays stable and shareable. Continuous measurement needs repeated runs, which is what Eli does once connected.

Why are the sources different from what ranks on Google?

AI systems with live search pick sources per answer and often favour pages that compare, review or discuss options. The overlap with classic rankings is partial.

The suite

Related free tools

All free tools

A free check finds the problem. Eli keeps fixing it.

Run the free check first. If the gaps matter, connect Eli and let it keep working on them.

Connect Eli