UI vs API

Genwolf collects AI responses using official model APIs.

This is a deliberate choice focused on measuring the core signals that actually determine AI visibility.

The layers behind an AI answer

An AI chat response is not a single system.
It's a stack of layers:

1

Model knowledge

What the model has learned and internalized.

2

External sources & citations

Which documents, domains, or references are pulled into the answer.

3

Chat software & personalization

UI logic, conversation history, location, user profiling.

The lower the layer, the more fundamental and stable it is.

What Genwolf measures

Genwolf focuses on the two most important layers:

1. Model-level knowledge & recognition

Whether the model:

  • knows your brand
  • associates it with the right category
  • considers it relevant to a given question

This is the foundation of AI visibility.

If your brand is not present here, no UI layer can reliably surface it.

2. Source-level signals

Which sources are cited or referenced:

documentation
blogs
review sites
authoritative domains

Sources strongly influence:

  • which brands appear
  • how they are framed
  • who is compared against whom

Tracking sources reveals why visibility happens, not just that it happens.

Why the UI layer matters less for measurement

Chat interfaces add:

formatting
response ordering
personalization
conversation memory

These layers affect how an answer is shown — but not whether a brand is fundamentally recognized.

They introduce noise without changing the underlying signal.

For measurement purposes, this layer is:

  • volatile
  • user-specific
  • hard to normalize

Why API-based data is sufficient

API access allows Genwolf to:

query the model's core knowledge
observe source selection behavior
run consistent, repeatable experiments

While API responses are not a 1:1 user experience, they capture the heart of AI visibility:

what the model knows and what it chooses to reference.

Why UI scraping doesn't solve realism

UI scraping:

  • still lacks real user context
  • often runs in clean or private sessions
  • repeats prompts mechanically

It attempts to reach the UI layer, but still misses true personalization.

UI scraping is also often unstable and can violate platform terms of service.

The result is higher complexity without access to more fundamental signals.

Trade-offs & limitations

What Genwolf captures well

  • model recognition
  • brand associations
  • source-driven visibility
  • relative competitive presence

What Genwolf does not capture

  • individual personalization
  • UI-specific ordering
  • session-based behavior

These limits are explicit and intentional.

Summary

AI visibility is determined bottom-up:

Model knowledgeSourcesUI & personalization

Genwolf measures the layers that matter most.

Not the surface. The core.