FAIV™FAIV™
Dimension08 / 08

Research

Metrics under active development

The behavior of generative AI models opens measurement dimensions that do not yet have a standardized methodology. Research metrics are concepts that FAIV™ has identified as relevant to brand AI visibility, and for which it is developing rigorous measurement frameworks. They are published at this stage to establish nomenclature and gather reference data.

PCI™KFI™AIS™
PCI™Research

Prompt Contextualization Index

CORE™ Module

AIS™ + SCAN™

Measures how your brand's presentation varies based on prompt context: whether AI adapts its description to user type, urgency or search intent.

AI models do not respond the same way to 'what health insurance do you recommend?' as to 'I need urgent health insurance for freelancers'. Prompt context — intent, urgency, user profile, explicit comparison — changes how AI presents options. PCI™ measures whether your brand benefits or is harmed by that contextual adaptation.

A brand can have excellent visibility in informational prompts ('what is X?', 'what are the options for Y?') and low visibility in transactional prompts ('I need to hire X today', 'what is the best option of Y for my case?'). PCI™ maps that distribution to identify in which types of prompts there is more work to do.

Real example

An e-learning platform has AIRI™ 52 (good overall visibility), but PCI™ reveals an important gap: it appears in 65% of informational prompts ('what are the best online training platforms?') but only in 18% of purchase prompts ('I need to learn Python this month, what platform should I use?'). AI knows it but does not recommend it when the user is ready to decide.

KFI™Research

Knowledge Freshness Index

CORE™ Module

SCAN™ + AIP™

Evaluates how current the information AI models have about your brand is, detecting gaps between your company's reality and the model's knowledge.

AI models have training cutoff dates and, although some incorporate real-time search, most of their brand knowledge is built in the training process. KFI™ measures the currency of that knowledge: does AI know about your rebrand from last year? About your new product line? About the change in your value proposition?

KFI™ is especially critical for companies that have had significant changes in the last 2-3 years: mergers, acquisitions, strategic pivots, geographic expansion, name change or brand repositioning. In these cases, AI may be communicating an old version of the company that no longer exists.

Real example

A Spanish financial technology company completed a rebrand in 2024: new name, new visual identity and new positioning ('from payment processor to financial intelligence platform'). Six months after the rebrand, all AI models are still using the old name, the old value proposition and describing services that were already discontinued. KFI™: 12/100.

AIS™Research

AI Sentiment Score

CORE™ Module

SCAN™ + AIS™

Analyzes the emotional and evaluative tone AI models use when referring to your brand: whether the language is positive, neutral, ambivalent or presents systematic reservations.

AI models are not emotionally neutral when talking about brands. While they try to be balanced, the training corpus incorporates the general tone with which sources talk about each company — and that tone is reflected in responses. AIS™ (AI Sentiment Score) analyzes whether that tone benefits or harms you.

Negative sentiment signals are often subtle: phrases like 'although some users report...', 'while it is an option...', 'it is worth verifying...', 'for certain profiles it may work...' are not explicit criticisms, but accumulated systematically they build a perception of second choice. AIS™ detects these patterns before they become critical.

Note: AIS™ as an AI visibility metric is different from the AIS™ (AI Strategy) module in CORE™. The AIS™ module is the solution; the AIS™ metric is the indicator that solution helps improve.

Real example

A financial technology company discovers through AIS™ that AI models always cite it with implicit reservations: 'although integration may require technical assistance', 'while it has a learning curve', 'for companies with sufficient IT resources'. No comment is incorrect — but the cumulative pattern positions the company as a complex solution, when its value proposition is precisely the simplicity of implementation.

Reference · Scale 0–100

0–25

Invisible

AI barely recognizes you in this dimension. Starting point for any visibility strategy.

26–50

Emerging

Incipient presence, very inconsistent across engines and contexts. AI is starting to pick up some signals.

51–75

Established

Solid presence. AI knows you in most relevant contexts and includes you regularly.

76–100

Reference

AI positions you as an authority source in your category. Spontaneous mention in most relevant responses.

How are you scoring in research?

Measure it with PULSE™ or SCAN™.

PULSE™ gives you your overall visibility score in minutes. SCAN™ audits research and all 7 other dimensions in depth.